Computer program, game system used for the same, and control method

Abstract

To provide a computer program capable of providing superiority information from various view points for each option in a game including a selection opportunity for selecting part of a plurality of options.SOLUTION: A computer program PG2 causes a control unit 31 incorporated into a user device 3 to function as the undermentioned means in a card game including a selection opportunity for selecting part of a plurality of options. Namely, the computer program PG2 causes the control unit 31 to function as: means for acquiring information on a plurality of probability values using a plurality of types of analysis model parts 23 (learned artificial intelligence models) for outputting information on a probability value (superiority information) for at least one option respectively for each option; and means for providing information on the probability values of part of the options for each option before selecting them.SELECTED DRAWING: Figure 5

Description

Technical Field

[0001] The present invention relates to a computer program or the like applied to a computer incorporated in a game system that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence exerted by each option before selecting at least one of a plurality of options in a game including an opportunity to select a part of the plurality of options.

Background Art

[0002] There is a game system that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence exerted by each option before selecting at least one of a plurality of options in a game including an opportunity to select a part of the plurality of options. For example, as a game, options, and superiority / inferiority information, a system is known in which shogi in the real space, a move in the shogi, and information such as "winning rate" for each move are respectively adopted (see, for example, Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Non-Patent Document 1 discloses a technique of adding information such as "winning rate", "candidate moves", and "reading lines" simulated for each move by AI (artificial intelligence) to a relay screen obtained by photographing the shogi board during a game from above. However, this information (simulation result) is only a simulation result based on one type of AI.

[0005] Generally, AI tends to have thoughts (or perhaps it could be called personality) influenced by learning datasets and the like that form the basis of machine learning. For example, due to differences in the content of the learning dataset or the learning method (algorithm) of machine learning, etc., AI often forms its own unique thoughts. Therefore, if the same move in a shogi game in Non-Patent Document 1 is simulated by multiple types of AI, there is a possibility that the simulation result of one AI may differ from that of another AI. When only the simulation result based on one type of AI is provided, the simulation result may be biased, or the simulation may be executed for a purpose different from the intention of the move. As a result, the reliability of the simulation result may decrease, or otherwise, a situation may arise where AI with different thoughts, in other words, simulation results from various perspectives are widely demanded. The simulation result is not limited to AI and may also be calculated by various other logics, but the same situation can occur in that case as well.

[0006] Therefore, an object of the present invention is to provide a computer program or the like that can provide superiority and inferiority information from various perspectives for each option in a game including a selection opportunity for selecting a part of a plurality of options.

Means for Solving the Problem

[0007] The computer program of the present invention is configured to function as a superiority and inferiority acquisition means for acquiring a plurality of superiority and inferiority information for each option of at least one of the plurality of options by using a plurality of types of processes for outputting the superiority and inferiority information for each of the at least one option to a computer incorporated in a game system that provides the user with the superiority and inferiority information regarding the superiority and inferiority of the influence exerted by each option in a game including a selection opportunity for selecting a part of a plurality of options before the selection of the part of the options, and a superiority and inferiority providing means for providing the plurality of superiority and inferiority information for each option before the selection of the part of the options.

[0008] On the other hand, the game system of the present invention is a game system incorporating a computer that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence of each option for at least one of a plurality of options in a game including a selection opportunity in which some of the options must be selected, before the user selects one of the options, and the computer functions as a superiority / inferiority acquisition means that acquires multiple pieces of superiority / inferiority information for each option of the at least one option using multiple types of processing that outputs the superiority / inferiority information for each of the at least one option, and a superiority / inferiority providing means that provides the multiple pieces of superiority / inferiority information for each option before the user selects the one of the options.

[0009] Furthermore, the control method of the present invention is a method for causing a computer incorporated in a game system that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence of each option for at least one of a plurality of options in a game including a selection opportunity in which some of the options must be selected, before the user selects one of the options, to execute the following steps: obtaining multiple pieces of superiority / inferiority information for each option of the at least one option using multiple types of processing that outputs the superiority / inferiority information for each of the at least one option; and providing the multiple pieces of superiority / inferiority information for each option before the user selects the some of the options. [Brief explanation of the drawings]

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Embodiments for Carrying Out the Invention

[0011] (Overall Configuration) Hereinafter, with reference to the drawings, a control method according to an embodiment of the present invention, a game system in which a computer program is implemented (a game system according to an embodiment of the present invention), etc. will be described. First, with reference to FIG. 1, the overall configuration of a network system to which the game system according to an embodiment of the present invention is applied will be described. As shown in FIG. 1, the network system 1 is configured as a client-server type system including a plurality of user devices 3 as clients and a game server 2 connected to each user device 3 via a network NT.

[0012] The user device 3 is a device for daily use by the user, and is a computer device (information communication terminal device) having an information communication function via the network NT. The user device 3 implements a computer program according to an aspect of the present invention, and functions as a game system according to an aspect of the present invention in the network system 1. As an example, a smartphone having a communication call function or a tablet terminal may be used as the user device 3. The user device 3 may be a PC (abbreviation for personal computer), or may be a personal or household game machine provided as a so-called consumer game machine. Further, as the user device 3, a business game machine provided as a so-called arcade game machine may be used.

[0013] The user device 3 functions as a game machine by implementing a predetermined software (application) and provides a game. The user device 3 provides a game including a selection opportunity for the user to select a part of a plurality of options. Such a game may be configured as an appropriate game such as a role-playing game, a simulation game, or an action game. As an example, it is configured as a one-on-one video game in which the user of the user device 3 and an opponent (another user including a computer) play against each other according to a predetermined rule through a game screen including a plurality of objects. In a one-on-one game, the user and the opponent may function as a member of a team, and the one-on-one game may be played in a one-to-many (including the case where a plurality of users other than oneself become independent opponents as in mahjong) or many-to-many format. Hereinafter, the case of playing in a one-on-one format will be described as an example.

[0014] In addition, the competitive game may be provided as a perfect information game such as shogi or chess (both are games of the type that use pieces as objects), but as an example, it is provided as an imperfect information game. For a perfect information game of the type in which all the contents of various elements necessary for selection in a selection opportunity are shared with the user, an imperfect information game is a game in which the contents of at least one element of a group of elements (for example, the contents of cards in a card game) used by other users (including computers) are not shared with the user. Various games such as mahjong are included in the imperfect information game. Hereinafter, a case where a card game is provided as an example thereof will be described.

[0015] The game server 2 may be configured by appropriately combining a plurality of server units (server devices), or may be configured by a single server unit. The game server 2 may be configured as a cloud server using cloud computing technology. The game server 2 provides various services related to the game to the user device 3. This service includes an analysis service and a speculation service. Both the analysis service and the speculation service are services for assisting selection in the selection opportunity of the game. Specifically, the analysis service is configured to provide the user with at least one option before selection with the superiority information regarding the superiority or inferiority of the influence given by each option included in the selection opportunity as an analysis result. On the other hand, the speculation service is configured to provide speculation information regarding information not shared with the user in the card game. Details of the analysis service and the speculation service will be described later.

[0016] Note that the game server 2 may also provide the user device 3 with, for example, a distribution service for distributing a computer program and various data necessary for the user device 3 to play a game, a matching service for matching users who cooperate or compete in the game, a relay service for relaying game information to be shared among the user devices 3, and the like.

[0017] The network NT may be configured as appropriate as long as it can connect the user device 3 to the game server 2. As an example, the network NT is configured to realize network communication using the TCP / IP protocol. Typically, the network NT is configured by combining the Internet as a WAN and an intranet as a LAN. In the example of FIG. 1, the game server 2 is connected to the network NT via the router NTr, and the user device 3 is connected to the network NT via the access point PP, respectively.

[0018] (Control System of Network System) Next, the main part of the control system of the network system 1 will be described with reference to FIG. 2. First, the game server 2 is provided with a control unit 21 and a storage unit 22 as a storage means. The control unit 21 is configured as a computer that combines a processor unit that executes various arithmetic processes and operation controls according to a predetermined computer program, and an internal memory and other peripheral devices necessary for its operation. The processor unit may appropriately include units such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), and an NPU (Neural network Processing Unit) (including cases where each processor unit is integrated as appropriate, such as when the GPU is incorporated into the CPU).

[0019] The storage unit 22 is an external storage device realized by a storage unit including a non-volatile storage medium such as a hard disk array (a computer-readable storage medium). The storage unit 22 may be configured to hold all data on a single storage unit, or may be configured to store data distributed across a plurality of storage units. The storage unit 22 records the server program PG1 and the server data SD. The server program PG1 is a computer program that causes the control unit 21 to execute processes necessary for providing various services to the user device 3. The server program PG1 may appropriately include various programs according to the processes to be executed by the control unit 21, and in the example of FIG. 2, includes the analysis program AP and the inference program FP.

[0020] The analysis program AP is a computer program that causes the control unit 21 to execute various processes for realizing an analysis service. The analysis program AP may be appropriately configured. For example, it may be configured to cause the control unit 21 to execute a process of outputting an analysis result based on a predetermined logic (including predetermined rules such as rules). In the example of FIG. 2, however, it is configured to cause the control unit 21 (e.g., GPU) to function as an artificial intelligence model that outputs an analysis result. The analysis program AP may be appropriately configured, but in the example of FIG. 2, it is configured as an inference program incorporating the learned data (parameters) AD for analysis output by causing the learning data set (for analysis) to be learned by the learning program (for analysis).

[0021] The learned data AD for analysis may be of one type, but in the example of FIG. 2, it includes multiple types of learned data AD. The multiple types of learned data AD for analysis may be generated by applying the same learning data set to learning programs with different algorithms (learning methods), but as an example, they are generated by applying different learning data sets to the same learning program. The analysis program AP functions the control unit 21 as an artificial intelligence model with different thoughts by incorporating and executing each of the multiple types of learned data AD for analysis. The analysis programs AP incorporating each of the multiple types of learned data AD for analysis are distinguished as the first analysis program AP1, the second analysis program AP2, etc. in the example of FIG. 2. For this reason, the analysis program AP is provided with a plurality of analysis programs AP such as these first analysis program AP1 and second analysis program AP2.

[0022] The inference program FP is a computer program for causing the control unit 21 to execute various processes for realizing an inference service. The inference program FP may be configured appropriately. For example, it may be configured to cause the control unit 21 to execute a process of outputting an inference result based on a predetermined logic. However, in the example of FIG. 2, it is configured to function the control unit 21 (e.g., GPU) as an artificial intelligence model, similar to the analysis program AP. The inference program FP may also be configured appropriately, but in the example of FIG. 2, it is configured as an inference program incorporating the learned data (parameters) FD for inference output by causing a learning data set (for inference) to be learned by a learning program (for inference).

[0023] The learned data FD for speculation may be of one type, but in the example of FIG. 2, it includes a plurality of types of learned data FD for speculation (not necessarily matching the number of learned data AD for analysis). The plurality of types of learned data FD for speculation may be generated by applying the same learning dataset to learning programs with different algorithms (learning methods), but as an example, they are generated by applying different learning datasets to the same learning program. The speculation program FP functions the control unit 21 as an artificial intelligence model with different thoughts by incorporating and executing each of the plurality of types of learned data FD for speculation. The speculation programs FP incorporating each of the plurality of types of learned data FD for speculation are distinguished as the first speculation program FP1, the second speculation program FP2, etc. in the example of FIG. 2. For this reason, the speculation program FP is provided with a plurality of speculation programs FP such as these first speculation program FP1 and second speculation program FP2.

[0024] The server data SD is data referred to by the server program PG1 to provide various services. The server data can include appropriate data related to various services, but in the example of FIG. 2, as an example, play data PD, learned data AD for analysis, and learned data FD for speculation are shown. The play data PD is data in which information regarding the past play results of each user is described. The play data PD may include not only play results but also other information necessary for managing each user, such as various personal information including the gender or address of each user. The possessions of each user within the card game can be managed as appropriate, but as an example, they are managed by the play data PD. Specifically, the card game is configured to grant a card object to a user who satisfies a predetermined granting condition (such as purchase, lending, lottery, or granting due to game progress). Then, the card objects granted to each user (hereinafter sometimes referred to as owned cards) are managed so as to be associated with each user in the play data PD.

[0025] The control unit 21 can be provided with various logical devices by combining the hardware resources of the control unit 21 and the server program PG1 as software resources. In the example of FIG. 2, an analysis model unit 23 and a speculation model unit 24 are provided. Both the analysis model unit 23 and the speculation model unit 24 are logical devices that function as artificial intelligence models. Specifically, the analysis model unit 23 functions as an artificial intelligence model that outputs analysis results based on the analysis program AP, and the speculation model unit 24 functions as an artificial intelligence model that outputs speculation results based on the speculation program FP.

[0026] The analysis model unit 23 (artificial intelligence model) is realized, for example, by a combination of the learned data AD for analysis and the analysis program AP (the analysis program AP that incorporates the learned data AD for analysis) and executes various processes for realizing the analysis service. This type of artificial intelligence model can form different thoughts depending on various elements. For example, it tends to have different thoughts due to at least one difference among the content of the learning data set, the algorithm of the learning program, and the algorithm of the inference program. The analysis model unit 23 can be provided with an appropriate type of analysis model unit 23 according to these differences. In the example of FIG. 2, two types of analysis model units 23, namely, a first analysis model unit 23A and a second analysis model unit 23B, are shown.

[0027] The first analysis model unit 23A is an analysis model unit 23 corresponding to the first analysis program AP1. Similarly, the second analysis model unit 23B is an analysis model unit 23 corresponding to the second analysis program AP2. The first analysis model unit 23A and the second analysis model unit 23B execute various processes according to the first analysis program AP1 or the second analysis program AP2. For example, the process includes option analysis processing. The details of the procedure of the option analysis processing will be described later.

[0028] Similarly, the estimation model unit 24 (artificial intelligence model) is realized by, for example, a combination of pre-trained data for estimation FD and an estimation program FP (estimation program FP that incorporates pre-trained data for estimation FD) and executes various processes for realizing an estimation service. An appropriate type (including one type) of estimation model unit 24 can be provided for the estimation model unit 24. In the example of FIG. 2, two types of estimation model units 24, i.e., a first estimation model unit 24A and a second estimation model unit 24B, are shown.

[0029] The first estimation model unit 24A is an estimation model unit 24 corresponding to the first estimation program FP1. Similarly, the second estimation model unit 24B is an estimation model unit 24 corresponding to the second estimation program FP2. The first estimation model unit 24A and the second estimation model unit 24B execute various processes according to the first estimation program FP1 or the second estimation program FP2. For example, the process includes a reverse side estimation process. Details of the procedure of the reverse side estimation process will be described later.

[0030] Note that, in the control unit 21, for example, a Web service management unit or the like for realizing processes related to various services such as the above-described distribution service, matching service, or relay service can be provided as a logical device. Similarly, an input device such as a keyboard and an output device such as a monitor can be connected to the control unit 21 as needed. However, their illustrations are omitted.

[0031] On the other hand, the user device 3 is provided with a control unit 31 and a storage unit 32 as storage means. The control unit 31 is configured as a computer that combines a processor unit that executes various arithmetic processes and operation controls according to a predetermined computer program and an internal memory and other peripheral devices necessary for its operation. The processor unit can appropriately include units such as a CPU, a GPU, and an NPU (including cases where each processor unit is integrated as appropriate, such as when a GPU is incorporated into a CPU), similar to the game server 2.

[0032] The storage unit 32 is an external storage device realized by a storage unit including a non-volatile storage medium (a computer-readable storage medium) such as a hard disk or a semiconductor storage device. The game program PG2 and the game data GD are stored in the storage unit 32. The game program PG2 is a computer program that causes the control unit 21 to execute the processes necessary for the user device 3 to function as a game device. The game data GD is data referred to by the game program PG2 for providing the game. The analysis program AP or the speculation program FP may be appropriately provided to the user device 3 from the game server 2. In that case, the game data GD may include the analysis program AP or the speculation program FP.

[0033] The game data GD may appropriately include various types of data (including various tables) necessary for playing the game, such as image data for displaying various images for the game, or audio data for playing various sounds (including BGM such as music), or card data defining each card object. In the example of FIG. 2, play data PD is shown. The play data PD is provided from the game server 2 as needed and stored. Also, when the analysis program AP or the speculation program FP is provided to the user device 3 from the game server 2, the game data GD may include the analysis learned data AD and the speculation learned data FD.

[0034] In the control unit 31, a progress control unit 33, a data management unit 34, and a model processing unit 35 are provided as logical devices realized by the combination of the hardware resources of the control unit 31 and the game program PG2 as software resources.

[0035] The progress control unit 33 executes various processes necessary for the progress of the card game. The processes include those necessary to enjoy the services provided by the game server 2. For example, the progress control unit 33 executes processes related to matching with an opponent in cooperation with the web service management unit, reflects the play of the opponent during the game in its own progress, or vice versa, reflects its own play to the opponent. In addition, the progress control unit 33 also executes processes such as providing selection opportunities and switching between each turn and phase, which will be described later.

[0036] The data management unit 34 executes various processes related to the management of the game data GD. For example, when a card object is given to a user, the data management unit 34 executes a process to reflect the card object as an owned card in the play data PD.

[0037] The model processing unit 35 executes various processes necessary to provide the user of the user device 3 with analysis services and speculation services. The processes executed by the model processing unit 35 include those realized in cooperation with the analysis model unit 23 or the speculation model unit 24 of the game server 2. For example, the model processing unit 35 executes a process to analyze the influence of at least one of a plurality of selectable options in a selection opportunity on the progress of the card game in cooperation with the analysis model unit 23 of the game server 2 and provide the result as an analysis result. Similarly, the model processing unit 35 executes a process to provide speculation information regarding information not shared with the user in the card game in cooperation with the speculation model unit 24 of the game server 2.

[0038] For example, as a process to realize the analysis service, the model processing unit 35 executes option analysis processing in cooperation with the analysis model unit 23 of the game server 2. In addition, the model processing unit 35 also executes display change processing. Similarly, as a process to realize the speculation service, the model processing unit 35 executes backside speculation processing in cooperation with the speculation model unit 24 of the game server 2. The details of the procedure of the display change processing will be described later.

[0039] In addition, the model processing unit 35 generates a situation log SL necessary for the analysis by the analysis model unit 23 or the speculation by the speculation model unit 24, causes the internal storage device of the control unit 31 to hold it, and also executes a process of appropriately updating the situation log SL as the situation changes. The situation log SL may appropriately include various information regarding the situation of the game. The situation log SL (data) includes, for example, information on all card objects currently in use, information on each card object arranged in various arrangement locations described later, attributes of each card object (including various states of each card object such as summoning described later or leaving the field area), life points, phases, and information on selectable options in each situation, and includes various information (which may be limited to an appropriate part of the information) for determining the situation of the card game.

[0040] The user device 3 is provided with appropriate output devices and input devices. In the example of FIG. 2, a monitor MO, a speaker SK, and a touch sensor TS are shown as examples of the output devices and the like. All of them are general-purpose hardware provided in an information communication terminal such as a smartphone. For example, the touch sensor TS is an input device that inputs a signal corresponding to a touch operation (an operation of touching with a finger) of the user to the control unit 31. The speaker SK is an output device for reproducing various sounds in response to a signal from the control unit 31. The monitor MO is an output device (display device) for presenting a game screen or the like in response to a signal from the control unit 31. Note that the user device 3 may be appropriately provided with various devices such as a gyro sensor, an acceleration sensor, and a position information (for example, GPS information) receiving device.

[0041] (Overview of the game) Next, the overview of the card game will be described with reference to FIGS. 3 to 11. The card game may be configured as an appropriate game, but as an example, it is configured as a video game played via a game screen including a plurality of card objects. More specifically, the card game is configured to be played using a deck of card objects (a group of a predetermined number of card objects) selected by the user for play from the owned card group.

[0042] FIG. 3 is a diagram schematically illustrating an example of a battle screen for playing a card game. The battle screen 50 is a game screen displayed when a user and an opponent battle using their respective deck cards. Various card objects may be arranged (displayed) on the battle screen 50. However, in the example of FIG. 3, for ease of explanation, regardless of type or content, card objects CO facing up (the direction in which the contents of the card object are visible) are shown with a dotted pattern, and card objects facing down (the direction in which the contents of the card object are not visible and are concealed) are shown with a diagonal right-hand line. In this example, a predetermined number of card objects CO (which may be constant or variable, and may not match the number of deck cards in the opponent's deck) included in the user's deck card function as the plurality of objects and card objects of the present invention. Furthermore, the owned card group and the deck cards function as the denominator element group and two or more elements of the present invention, respectively.

[0043] As shown in Fig. 3, the battle screen 50 includes a user area 51, a common area 52, and an opponent area 53. The user area 51 is an area dedicated to the user. The user area 51 can be configured as appropriate, but in the example of Fig. 3, it includes a field area 54, a main deck placement area 55, an extra deck placement area 56, a hand placement area 57, a graveyard area 58, and a point display area LR.

[0044] The field area 54 is an area where a plurality of card placement areas CP (only some are labeled) are formed. Each card placement area CP is a place (area) where each card object CO for advancing the game among the deck cards should be placed. In the card game, appropriate types of card objects CO with various roles (uses), effects, etc. can be prepared. For example, a card object CO representing a monster (hereinafter sometimes referred to as "monster card CO1"), a card object CO representing a magic (predetermined effect) (hereinafter sometimes referred to as "magic card CO2"), a card object CO representing a trap (an effect different from magic) (hereinafter sometimes referred to as "trap card CO3"), or a card object CO representing a special effect (hereinafter sometimes referred to as "special card CO4") and the like are prepared. When a card object CO is placed in each card placement area CP, that card object CO actually exerts its role and the like in the battle.

[0045] For example, the monster card CO1 is given roles such as the opponent in a battle, attacking or defending against the opponent's monster card CO1. Also, the monster card CO1 can be classified into various types. As an example, it includes two types: the normal monster card CO1 and the extra monster card CO1. The normal monster card CO1 is a monster card CO1 that can be placed from the hand placement area 57 to the field area 54. On the other hand, the extra monster card CO1 is a monster card CO1 that is summoned (called) in exchange for the monster card CO1 leaving the field area 54. The placement (summoning) of the extra monster card CO1 is usually restricted to the shared area 52, but it may be allowed to be placed in the field area 54 when certain field conditions are met. For this reason, the monster card CO1 may have the role of summoning (calling) other monster cards CO1 to the field area 54, etc. And the monster card CO1 placed in the card placement area CP is used to actually give roles such as attack, defense, or summoning to the progress of the game. Similarly, the magic card CO2, the trap card CO3, and the special card CO4 are card objects CO that can be placed from the hand placement area 57 to the field area 54, and the magic card CO2, the trap card CO3, and the special card CO4 placed in the card placement area CP are used for the activation of the effects indicated by each card object CO.

[0046] A plurality of card placement areas CP may be appropriately placed in the field area 54. In the example of FIG. 3, they are arranged to form two rows, the front row 54A located closer to the opponent's area 53 and the rear row 54B located behind it. The front row 54A and the rear row 54B are rows formed by six card placement areas CP and five card placement areas CP arranged side by side, respectively. Any card object CO can be placed in each card placement area CP without restriction, but as an example, there are restrictions on the types of card objects CO that can be placed. The front row 54A and the rear row 54B may be appropriately used separately, but as an example, they are used separately according to the types of card objects CO that can be placed.

[0047] Specifically, the front row 54A is used as each card placement area CP where the monster card CO1 among the types of card objects CO should be placed. However, only the leftmost card placement area CP (which can be formed as appropriate, but in the example of FIG. 3, is formed slightly smaller than the other card placement areas CP) among the six card placement areas CP forming the front row 54A is an exception and is used as each card placement area CP where the special card should be placed. That is, the placement of card objects CO such as the magic card CO2 or the trap card CO3 on the front row 54A is restricted, the placement of the special card CO4 on the front row 54A other than the leftmost card placement area CP is restricted, and the placement of the monster card CO1 is further restricted at the leftmost card placement area CP.

[0048] On the other hand, the rear row 54B is used as each card placement area CP where the magic card CO2 and the trap card CO3 among the types of card objects CO should be placed. That is, the placement of card objects CO such as the monster card CO1 or the special card CO4 on the rear row 54B is restricted. Note that a monster card CO1 having the characteristics of the magic card CO2 (the monster card CO1 corresponding to a monster having a magic effect) may be prepared, and the placement of such a monster card CO1 having a magic attribute on a predetermined card placement area CP (for example, the card placement areas CP at both the left and right ends) of the rear row 54B may be permitted. That is, it may include card objects CO having a plurality of appropriate attributes, and the placement of such card objects CO may be permitted in both the front row 54A and the rear row 54B according to the plurality of attributes.

[0049] In the field area 54, under the restrictions for each card placement area CP, card objects CO can be appropriately arranged according to the user's instructions. In the example of FIG. 3, one monster card CO1 is arranged in the third card placement area CP from the right in the front row 54A. Also, each card object CO may be arranged in the card placement area CP to face various directions such as vertically or horizontally, and the arrangement directions such as vertical can be appropriately used as needed. As an example, when the monster card CO1 is used for defense, it is arranged horizontally, and when it is used for attack, it is arranged vertically, respectively.

[0050] The hand arrangement area 57 is an area where card objects CO (hereinafter sometimes referred to as hand CO5) that are virtually in hand among the deck cards should be arranged (displayed). The hand CO5 is a candidate card object CO to be arranged in the field area 54 (each card placement area CP). An appropriate number of hand CO5 can be arranged in the hand arrangement area 57. In the example of FIG. 3, five hand CO5 are arranged. In this example, the hand arrangement area 57 functions as the hand area of the present invention.

[0051] The main deck placement area 55 and the extra deck placement area 56 are both areas where card objects CO indicating the remaining deck cards (the remainder after excluding the card objects CO in the field area 54 and the hand placement area 57 from the initial deck cards) should be placed, but their uses are different. Specifically, the main deck placement area 55 is the area where card objects CO that are candidates to be added to the hand CO5 among the remaining deck cards (hereinafter sometimes referred to as the deck CO6) should be placed. The card objects CO of the deck CO6 are added to the hand CO5 through draws in the draw phase described later. On the other hand, the extra deck placement area 56 is the area where a group of card objects (hereinafter sometimes referred to as the summoning deck card CO7) as a bundle of extra monster card CO1 should be placed. When the summoning of the extra monster card CO1 is executed via a normal monster card CO1, the extra monster card CO1 to be placed in the field area 54 is drawn from the summoning deck card CO7. That is, the remaining deck cards are divided into the deck CO6 and the summoning deck card CO7 and placed in the main deck placement area 55 and the extra deck placement area 56 respectively. In this example, the main deck placement area 55 functions as the deck area of the present invention.

[0052] The Graveyard Zone 58 is, in principle, an area for accommodating card objects CO that have left the Field Zone 54. For example, the Monster Card CO1 leaves the Field Zone 54 when it meets certain conditions (conditions for leaving the Field Zone 54 etc.), such as being attacked by the opponent's Monster Card CO1 or being Summoned as an Extra Monster Card CO1. Similarly, card objects CO that exhibit certain effects, such as the Spell Card CO2 and the Trap Card CO3, also leave the Field Zone 54 according to the card effects after their effects are activated. The Graveyard Zone 58 is provided as a place (destination of movement) for accommodating (placing) the card objects CO that have left the Field Zone 54. The Graveyard Zone 58 may be configured to display the card objects CO that have left the Field Zone 54, but in the example of Fig. 3, the display of the card objects CO is configured to be omitted.

[0053] The Point Display Area LR is an area for displaying the user's Life Points. In the card game, appropriate Life Points can be set for both the user and the opponent, but in the example of Fig. 3, all Life Points are set to 8000 points ("8000"). The win / loss conditions for determining the win or loss in the battle can be set as appropriate, but for example, they are satisfied when the opponent's Life Points reach zero. That is, the Life Points function as a parameter for determining the win or loss. Specifically, the user can reduce the opponent's Life Points by using the Monster Card CO1 placed in the Field Zone 54 and the Shared Zone 52 for an attack. If the opponent's Life Points can be reduced to zero, the user is determined to have won and the battle ends. On the other hand, if the user's Life Points reach zero due to an attack from the opponent's Monster Card CO1, the user is determined to have lost and the battle ends.

[0054] The shared area 52 is an area shared by the user and the opponent. The shared area 52 can be configured as appropriate. In the example of FIG. 3, two card placement areas CP are provided. In each card placement area CP, the user's card object CO and the opponent's card object CO are appropriately arranged. The arrangement of each card placement area CP in the shared area 52 may be allowed without limitation regardless of the type of the card object CO, but is limited to the extra monster card CO1 as an example. That is, the extra monster card CO1 summoned from the summoning deck card CO7 is first arranged in each card placement area CP of the shared area 52, and the arrangement in the field area 54 is allowed only when a predetermined field condition is satisfied. In this example, each card placement area CP (the extra deck placement area 56 may be included) of the user area 51 and the shared area 52 functions as a plurality of card placement areas of the present invention.

[0055] The opponent area 53 is an area dedicated to the opponent. The opponent area 53 is an area that plays the same role as the user area 51 for the opponent. Therefore, a field area 54, a main deck placement area 55, an extra deck placement area 56, a hand placement area 57, a graveyard area 58, and a point display area LR are also formed in the opponent area 53. However, since their roles and the like are the same as those of the user area 51, the description thereof is omitted.

[0056] FIG. 4 is a diagram showing an example of the progress procedure of the card game in the example of FIG. 3. As shown in FIG. 4, the card game includes the user's turn and the turn of the opponent (hereinafter, may be referred to as a player when not distinguishing between the two). And through these turns, it proceeds in a so-called turn-based system where the turns between the user and the opponent are alternately repeated. Specifically, first, as game preparation, for example, the player's deck cards are shuffled and then placed in the main deck placement area 55, and a predetermined number of card objects CO are drawn (drawn) from each player's deck cards (the deck CO6) and displayed (placed) in the hand placement area 57 as the hand CO5. Such processing is performed. When the preparation is completed, the game starts from the turn of the leading player (as an example, the user is the leading player). One turn is divided into a plurality of phases. A phase is a concept for dividing the procedure to be performed in one turn into multiple stages according to its content and nature. In the example of FIG. 4, one turn is divided into 6 stages from the draw phase to the end phase, but this is only an example.

[0057] In each phase, the player whose turn it is can select appropriate actions within the range defined for each phase. An example is as follows. In the draw phase, a card object CO is drawn from the deck CO6. In the standby phase, the effects of the card objects CO designated as those for which effect processing should be performed in that phase can be activated. In the first main phase, various actions such as the summoning of various objects such as monsters to be used in battle, the setting of card objects CO with unique effects such as spells and traps, or the activation of effects are permitted while appropriately using the card objects CO in the field area 54. In the battle phase, battles (battles) using card objects CO are conducted. For example, a battle is conducted by selecting a monster card CO1 to be used in the user's turn's attack and a monster card CO1 to be the target of the opponent's attack (if there is no monster card CO1 in the field area 54, a direct attack on the life points). The result of the battle is determined according to parameters such as the attributes and strength of the monster card CO1. In the second main phase, the same actions as in the first main phase are permitted. In the end phase, the end of the turn is indicated (declared).

[0058] Note that the battle in the battle phase can be avoided by the choice of the player whose turn it is. In that case, the battle phase and the second main phase are skipped. Similarly, in the battle phase, the second main phase can also be avoided by the player's choice. The end of one phase is indicated by a predetermined end operation. When one turn ends, the turn passes to the opponent player. When a predetermined win-loss condition is met while the turns are being repeated alternately, the game ends. As an example, the win-loss condition is met when the life points set for each player P are reduced to a predetermined value (for example, 0) by battle.

[0059] (Analysis service) Next, with reference to FIG. 5, the details of the analysis service will be described. The analysis service is a service that provides superiority and inferiority information as an analysis result for at least one option in the selection opportunity as described above. The analysis service can be applied to an appropriate situation including the selection opportunity, for example, applied to the selection opportunity included in the battle screen 50. FIG. 5 schematically shows an example of the battle screen 50 when the analysis service is applied. In the example of FIG. 5, the same components as those of the battle screen 50 in FIG. 3 are denoted by the same reference numerals and their description is omitted.

[0060] The battle in the battle screen 50 is played by repeating turns including a plurality of phases between players as shown in FIG. 4. For this reason, a phase end instruction for ending each phase (for example, entering the next battle phase in the case of the first main phase) or a turn end instruction for ending each turn can function as an option on the battle screen 50. Also, in the battle screen 50, as described above, deck cards (card object groups) are used through each phase and are arranged in the field area 54 etc. via the hand CO5. For this reason, the user is required to make selections such as which hand CO5 to place in which card placement area CP and how. Selections such as at what timing and how to activate the effects for magic cards CO2 and trap cards CO3 are also required. Also, in the case of monster cards CO1, selections such as whether to attack in the battle phase and if so, which monster card CO1 to attack are also required. When the analysis service is applied to the battle screen 50, superiority and inferiority information regarding these options is provided.

[0061] Specifically, as shown in FIG. 5, when the analysis service is applied, an advice display column 60 is additionally displayed as compared with the example of FIG. 3. The advice display column 60 is a column for displaying superiority and inferiority information (analysis result) for each option provided by the analysis service. The presence or absence of use of the analysis service, in other words, the presence or absence of display of the advice display column 60 may be set as appropriate and may be displayed by a predetermined operation etc. for requesting display, but as an example, it is automatically displayed for each selection opportunity without a request from the user.

[0062] The advice display column 60 may display an appropriate number of superiority and inferiority information. For example, the superiority and inferiority information regarding all options may be displayed. However, in the example of FIG. 5, five pieces of advice information are displayed via five advice sections 61 from the first advice section 61A to the fifth advice section 61E. Also, in the analysis service, a plurality of artificial intelligence models provided in the game server 2 are used, and via them, a plurality of superiority and inferiority information (analysis results) are output for each option. Therefore, in the advice display column 60, a plurality of superiority and inferiority information output by a plurality of artificial intelligence models are displayed for each option. The number (type) of artificial intelligence models may be an appropriate number. However, in the example of FIG. 5, two types of artificial intelligence models, an "attack type AI" and a "defense type AI", are used, and two types of superiority and inferiority information are displayed for each option by them.

[0063] "Attack type AI" and "defense type AI" respectively indicate an attack type artificial intelligence model (hereinafter sometimes referred to as attack type AI) and a defense type artificial intelligence model (hereinafter sometimes referred to as defense type AI). And the five advice sections 61 of the advice display column 60 are classified into an advice section 61 that displays superiority and inferiority information based on the attack type AI and an advice section 61 that displays superiority and inferiority information based on the defense type AI. The advice section 61 corresponding to the attack type AI is displayed in white in the example of FIG. 5. For example, the first advice section 61A, the third advice section 61C, and the fourth advice section 61D correspond to it. On the other hand, the advice section 61 corresponding to the defense type AI is displayed in a gray color scheme in the example of FIG. 5. For example, the second advice section 61B and the fifth advice section 61E correspond to it.

[0064] Each artificial intelligence model is realized by the content of the analyzed pre-trained data AD as an example as described above, and each analyzed pre-trained data AD is generated by the content of the training dataset. Then, the attack-type AI corresponds to the pre-trained artificial intelligence model based on the analyzed pre-trained data AD generated by the attack-type training dataset, and the defense-type AI corresponds to the pre-trained artificial intelligence model based on the analyzed pre-trained data AD generated by the defense-type training dataset. For example, among the plurality of types of analysis model units 23, the first analysis unit 23A corresponds to the attack-type AI, and the second analysis model unit 23B corresponds to the defense-type AI. The attack-type training dataset and the defense-type training dataset can be configured appropriately. For example, they are respectively composed of a dataset including a performance set that won with an attack-emphasized strategy and a dataset including a performance set that won with a defense-emphasized strategy.

[0065] Each advice unit 61 can be configured appropriately. In the example of FIG. 5, it includes an advice information unit 62 and an option information unit 63. The advice information unit 62 is a part that displays detailed information including superiority and inferiority information. The superiority and inferiority information may be appropriate information regarding the influence of each option, for example, various information such as symbols indicating the degree of influence of each option on the victory of A, B, C, etc., or character information that sensuously indicates the influence given by each option such as "likely to lead to victory". In the example of FIG. 5, it is configured as numerical information that indicates the possibility of the progress being advantageous in the battle through the magnitude of the numerical value. For example, in the first advice unit 61A, numerical information of "42.0%" is displayed as the superiority and inferiority information. The numerical information may be a numerical value indicating various advantages. For example, it may be a numerical value indicating the possibility of progressing advantageously (being less likely to become disadvantageous) within a predetermined range of turns or within the selection of the analysis target. In the example of FIG. 5, it is displayed as a probability value indicating the possibility (probability) of finally winning in the battle.

[0066] The probability value is information indicating the winning rate when compared with other options output by the same artificial intelligence model. That is, for example, if it is the probability value of an attack-type AI model, it is set so that the sum of the probability values of all options output by the attack-type AI model becomes 100%. The numerical information may be, for example, values independent of each other such as the evaluation results for each option by the attack-type AI model (the sum of the values of all options does not necessarily become 100%), but as an example, it is calculated as a probability value that totals 100% in this way. In addition, the detailed information may include appropriate information in addition to the information of the probability value (superiority and inferiority information). In the example of FIG. 5, information indicating the artificial intelligence model that output the probability value, such as "Attack-type AI A-type evaluation", is included. Note that in the example of FIG. 5, only one type (for example, "A type") of both the attack-type AI and the defense-type AI is used. However, a plurality of the same type (such as attack-type) of artificial intelligence models based on different performance sets, such as "B type", may be used for outputting the probability value.

[0067] The option information section 63 is a section that displays information on the options that are the target of the superiority and inferiority information (probability value). For example, in the option information section 63 of the first advice section 61A, information on the option of "Enter the battle phase" (an option corresponding to the start of the battle phase) is displayed. In this case, the first advice section 61A indicates that the probability value (superiority and inferiority information) of the option of "Enter the battle phase" is "42.0%". Note that in the example of FIG. 5, the display of the advice information section 62 and the option information section 63 of the third advice section 61C to the fifth advice section 61E is simplified. However, for example, by a touch operation on each advice section 61, details may be displayed in the same way as the first advice section 61A and the like.

[0068] Also, in the example of FIG. 5, the probability value regarding the option of "entering the battle phase" is output only by the attack-type AI, but the probability values regarding the same option are also output by the defense-type AI. Specifically, both the attack-type AI and the defense-type AI output probability values for all the options that can be selected in each situation of the battle screen 50. Therefore, at least two types of probability values corresponding to the attack-type AI and the defense-type AI respectively are output for each selectable option. The advice display column 60 may be configured to display all the information of the two types of probability values regarding each option, but in the example of FIG. 5, it is configured to display only the probability values of the top five options (some options).

[0069] The ranking of the probability values is evaluated for each artificial intelligence model such as the attack-type AI and the defense-type AI, and the top of each may be displayed in the advice display column 60, but in the example of FIG. 5, it is evaluated across the attack-type AI and the defense-type AI (different artificial intelligence models). For example, although "entering the battle phase" is evaluated as having the highest probability value in the attack-type AI, if "entering the battle phase" is evaluated as having a low probability value in the defense-type AI, only the analysis result of the attack-type AI may be displayed in the advice display column 60. Thus, as the probability value regarding the same option, only one of the attack-type AI and the defense-type AI may be displayed in the advice display column 60 in some cases, but as described above, both the attack-type AI and the defense-type AI output the probability values regarding the same all options, and both of them can be displayed as advice in the advice display column 60.

[0070] (Method for calculating probability value) Next, with reference to FIGS. 6 and 7, a method for calculating the probability value (superiority / inferiority information) for each option output by each artificial intelligence model such as an attack-type AI will be described. FIG. 6 is an explanatory diagram for explaining a list of options (actions) that a player can take in a situation of the battle screen 50. In the example of FIG. 6, the options that the player can take (conceptualized by "○") are shown in a tree type in which each option branches into the next option. As shown in FIG. 6, the options that the player can take on the battle screen 50 are classified into layers of a plurality of option groups, and the option group in the deepest layer in each tree corresponds to each actually selectable option. Then, all the actually selectable options are input into a plurality of artificial intelligence models such as an attack-type AI and a defense-type AI, and a probability value for each option is output by each of the plurality of artificial intelligence models.

[0071] Specifically, in the example of FIG. 6, in the situation where three monster cards CO1 are already arranged in the front row 54A of the field area 54 in the example of FIG. 5, and three magic cards CO2 etc. are already arranged in the rear row 54B, and two of "A monster card" and "B magic card" are held as the hand cards CO5, the selectable option group is shown. In this case, the first options that the player can take are the use of the "A monster card", the use of the "B magic card", the "phase transition", and the "turn end". Therefore, these four options form the first layer option group that can be selected first.

[0072] In addition, for the use of "A Monster Card" and "B Magic Card", there are further options regarding where and how to use them, and these form the second-layer option group. For example, in the field area 54 of the example in FIG. 5, the front row 54A is distinguished from the left end (the card placement area CP where the Monster Card CO1 cannot be placed) in order from the adjacent position to the right as the 1st card placement area CP to the 5th card placement area CP. When three Monster Cards CO1 are placed in the 3rd card placement area CP to the 5th card placement area CP, the card placement areas CP where the "A Monster Card" (Monster Card CO1) can be placed are the 1st card placement area CP and the 2nd card placement area CP. Therefore, these two card placement areas CP correspond to the options for the placement destinations of the "A Monster Card". Also, there are two types of methods for placing the Monster Card CO1 in the card placement area CP: "Summon" (the Monster Card CO1 is placed face-up) and "Set" (the Monster Card CO1 is placed face-down). Therefore, for each of the two options of the 1st card placement area CP and the 2nd card placement area CP, there are further two types of options: "Summon" and "Set". As a result, for the use of the "A Monster Card", there are four options corresponding to two card placement areas CP × two types of placement methods. Note that for the Monster Card CO1, there may be various other options depending on the situation. For example, in the case of an Extra Monster Card CO1, it is necessary to specify a predetermined number (e.g., three, etc.) of Monster Cards CO1 that should be moved away from the field area 54 (should be moved to the Graveyard area 58) for its summoning. In this case, options such as a predetermined number (e.g., three) × the types of Extra Monster Cards CO1 × its card placement area CP may be derived.

[0073] On one hand, for example, in the field area 54 in the example of FIG. 5, the rear row 54B is distinguished from the 6th card placement area CP to the 10th card placement area CP in order from the left end to the right. When three magic cards CO2 etc. are arranged in the 8th card placement area CP to the 10th card placement area CP, the card placement areas CP where the "B magic card" (magic card CO2) can be placed are the 6th card placement area CP and the 7th card placement area CP, two in number. For this reason, these two card placement areas CP correspond to the options for the placement destination of the "B magic card". Also, the "B magic card" can be configured as an appropriate magic card CO2. For example, if the "B magic card" is a magic card CO2 having two types of placement methods, namely "activation" (option to activate the effect) and "set" (option to defer the activation of the effect) as the placement method to the card placement area CP, these two types of placement methods exist as options. As a result, there are also four options corresponding to two card placement areas CP × two types of placement methods for the use of the "B magic card". And when these four options are added to the four options corresponding to the "A monster card", a total of eight options form the second-layer option group.

[0074] When an effect is set for the "A monster card", further options are derived. For example, when a conditional effect such as placement by "summoning" to the card placement area CP is set for the "A monster card", there are further two types of options for the placement method of "summoning", namely "effect activation" and "no effect activation". For this reason, four options corresponding to these two types of options × two card placement areas CP are generated from the second-layer option group regarding the "A monster card".

[0075] Similarly, options also arise when two types of effects are set for the "B Magic Card". For example, when two types of effects, namely A effect and B effect, are set for the "B Magic Card", there are further two options for the placement method, namely "Activate A effect" and "Activate B effect". Therefore, four options corresponding to these two types of options × two card placement areas CP are generated from the second-layer option group related to the "B Magic Card". And adding the four options derived from the second-layer option group of the "A Monster Card" to these four options forms a total of eight options that form the third-layer option group. Note that for the Magic Card CO2, there can be various other options depending on the situation. For example, when the activation target of the effect can be specified, an option to select the target may further arise. Although it is omitted in the example of Figure 6, the same applies to other card objects CO such as the Trap Card CO3.

[0076] Furthermore, a fourth-layer option group is derived from the third-layer option group. For example, if two of the three Monster Cards CO1 already placed in the card placement area CP (for example, the "C Monster Card" and the "D Monster Card") can be the activation targets of the effect of the "B Magic Card", these two activation targets are derived as options. Therefore, eight options corresponding to these two activation targets × four options (two types of effects generated for each of the two card placement areas CP) form the fourth-layer option group.

[0077] In the example of FIG. 6, in the tree of option groups derived from the "A monster card", a part of the second-layer option group (when arranged in the "set" state in each card placement area CP), and the third-layer option group correspond to the deepest-layer option group. Therefore, each option of these option groups is input into an artificial intelligence model such as an attack-type AI, and a probability value (the numerical value within the "○") is output for each option. Similarly, in the tree of option groups derived from the "B magic card", a part of the second-layer option group (when arranged in the "set" state in each card placement area CP), and the fourth-layer option group correspond to the deepest-layer option group. Therefore, each option of these option groups is input into an artificial intelligence model such as an attack-type AI, and a probability value is output for each option.

[0078] On the other hand, there are no options derived from "phase transition" (phase end instruction) and "turn end" (turn end instruction), and for these, the first-layer option group corresponds to the deepest-layer option group. Therefore, the two options of "phase transition" and "turn end" are input into an artificial intelligence model such as an attack-type AI, and a probability value is output for each option. In the example of FIG. 6, as an example, the information of the probability values output by one type of artificial intelligence model among a plurality of artificial intelligence models is shown, but actually, for each option, the information of a plurality of probability values is output by each of the plurality of artificial intelligence models.

[0079] In the example of FIG. 6, an example of a group of options in the battle phase is shown. In addition to the phase end instruction and the like, a card object CO to be placed from the hand CO5 to each card placement area CP, and an option corresponding to the card placement area CP where the card object CO is to be placed are shown. However, for example, in the draw phase, options corresponding to the phase, such as adding (drawing) from the deck CO6 to the hand CO5, are prepared for each phase. Therefore, at least a part of the options that can be selected for each phase (for example, they may be common at the end of the phase or at the end of the turn) are different. That is, the options available in each phase are different. And information on the probability values of each option that can be selected for each phase is output. In this example, the group of options in the deepest layer functions as an example of the plurality of options of the present invention. Also, the group of options for each phase functions as two or more options of the present invention.

[0080] FIG. 7 is an explanatory diagram for explaining an example of a method by which an artificial intelligence model calculates probability values. Generally, an artificial intelligence model is generated by machine learning a learning dataset using a predetermined learning method (algorithm), and tends to have a thinking (algorithm) corresponding to the learning method. The learning methods in machine learning include various methods. For example, they include imitation learning and reinforcement learning (imitation learning may be considered a type of reinforcement learning in some cases). Imitation learning is often classified into methods such as Behavoir clonig, Dataset Aggregation, or Inverse Reinforcement Learning, but there are also many cases where the reward is not explicitly defined. On the other hand, reinforcement learning is often classified into methods such as Dynamic Programming (DP method, dynamic programming method), MC method (Monte Carlo method), or Temporal Differerence Learning (TD method), and is often a method that maximizes the reward.

[0081] An artificial intelligence model (analysis program AP) for analysis services may be appropriately generated by various learning methods. As an example, it is generated by the TD method among reinforcement learning methods. Further, the TD method may include techniques such as SARSA, etc. As an example, the Q-Learning (Q-learning) technique is used for generating the artificial intelligence model for analysis services.

[0082] Q-learning is a type of technique that evaluates actions (options) by obtaining an action value function (Q-function). The Q-function is generally defined as a function that predicts how the future reward (often called the Q-value) will be when a certain action is taken in a certain state (the description of the specific function formula is omitted). However, in Q-learning, when the Q-function table (input information) becomes large (enormous), the calculation cannot keep up, and it tends to be difficult to implement. On the other hand, for example, the input information (information of the situation log SL) on the battle screen 50 is considered to be enormous. Therefore, a learned artificial intelligence model for analysis services is generated by the DQN (Deep Q Network) technique that obtains an approximate value of the Q-value using a neural network. In this case, the artificial intelligence model for analysis services is configured to calculate a probability value (Q-value) using the DQN technique.

[0083] Also, the reward that is the target of the Q-value (a predetermined result obtained by an action, which may be included in the algorithm of the learning program) may be appropriately set. For example, specific winning methods such as winning by a narrow margin or trends of winning methods may be set. The artificial intelligence model for analysis services may have different thoughts depending on the reward (predetermined result) in Q-learning. As an example of the reward, victory in a battle is set, and the artificial intelligence model for analysis services is configured to calculate a probability value (Q-value) for winning in a battle using the DQN technique.

[0084] In the example of FIG. 7, an outline of the DQN method for calculating probability values (Q values) is shown. As shown in FIG. 7, in the DQN method, since a neural network (deep learning) is used, an input layer, an intermediate layer, and an output layer are formed, and each layer cooperates to output a calculation result for data. The input layer is a layer that is responsible for collecting data. The intermediate layer is a layer that performs calculations for calculating probability values. The intermediate layer often includes multiple layers (generally, the higher the number of layers, the higher the accuracy tends to be). In the example of FIG. 7, only two intermediate layers are shown, but an appropriate number of intermediate layers may be formed. The output layer is a layer that outputs the calculation result performed in the intermediate layer. The artificial intelligence model learned by the DQN method is configured to output a probability value of winning in a battle at the output layer. Also, between each of the input layer, the intermediate layer, and the output layer, connection lines (often called synapses) connecting the layers are provided. And each connection line is provided with a value of weight (often represented by the symbol of w) indicating importance, and the importance of information is determined by the magnitude of the weight value. In the example of FIG. 7, both the input value (input information) and the output value (sometimes generally called a node) are represented by "○".

[0085] Specifically, first, information of the situation log SL indicating the current state of the battle screen 50 is input to the input layer. The situation log SL includes information of many dimensions (input values). It is desirable that the number of information in this dimension is less than about 5000. The situation log SL may include various information for discriminating the situation of the card game as described above. For example, in addition to information on the placement situation in the field area 54 such as the situation of the 1st card placement area CP to the 10th card placement area CP, information on possible options such as turn end, summoning of A monster card, activation of A monster card effect, activation of B magic card (see, for example, the example of FIG. 6) is included.

[0086] A predetermined functional equation using appropriate weighting (weight value w) is applied to the input values in the input layer, and the output values of the intermediate layer (the first layer) are calculated by this functional equation. Although an appropriate number of output values can be calculated for the first-layer intermediate layer, in the example of FIG. 7, four output values are calculated. Further, a predetermined functional equation using appropriate weighting is applied to the four output values, and the output values of the intermediate layer (the second layer) are calculated by this functional equation. Although an appropriate number of output values can also be calculated for the second-layer intermediate layer, in the example of FIG. 7, four output values are still calculated. Note that the weighting (weight value w) corresponding to each connection line is calculated as a learning result and is managed, for example, in the analysis-use learned data AD. Also, the connection lines (synapses) connecting the input layer and the intermediate layer are appropriately omitted in FIG. 7.

[0087] The output values of the final intermediate layer (the second layer) are calculated as Q values by a predetermined functional equation using appropriate weighting in the output layer. The Q values (output values of the output layer) are calculated for each predetermined action (options on the battle screen 50). For this reason, for example, in each tree in the example of FIG. 6, Q values are calculated by the method in the example of FIG. 7 for each option (selectable option) in the option group of the deepest layer. Also, the calculated Q values are converted into probability values by a normalization function (for example, the Softmax function) so that the sum of the values of all selectable options becomes 100%. As an example, probability values are calculated for all selectable options for each situation by such a method. Also, the same calculation is executed by each of a plurality of artificial intelligence models such as an attack-type AI or a defense-type AI (for example, the first analysis program AP1 and the second analysis program AP2). Then, the top (for example, five in the example of FIG. 5) of the plurality of probability values calculated by the plurality of artificial intelligence models are provided to the user via the advice display column 60. Note that the application of the normalization function is not essential. For example, when various conditions such as the learning method are different, the application of the normalization function may be appropriately omitted.

[0088] (Inference Service) Next, with reference to FIG. 8, the details of the speculation service will be described. The speculation service is a service that provides speculation information for speculating on the content of elements of an opponent that are not shared with the user among a group of elements (for example, card object CO) prepared for each player in a card game. The speculation service can be applied to an appropriate situation including elements hidden from the user. For example, it is applied to the speculation of card objects CO such as card objects CO placed face-down in the hidden field area 54 on the battle screen 50, hand cards CO5, a deck of cards CO6, or a summoned deck card CO7. That is, on the battle screen 50, speculation information for speculating on the content of card objects CO placed face-down and the like is provided via the speculation service. FIG. 8 schematically shows an example of the battle screen 50 when the speculation service is applied. In the example of FIG. 8, the same components as those of the battle screen 50 in FIG. 3 are denoted by the same reference numerals and their description is omitted.

[0089] As shown in FIG. 8, when the speculation service is applied, a speculation information section 70 is additionally displayed on the battle screen 50 as compared with the example of FIG. 3. The speculation information section 70 is a part for displaying speculation information. The speculation information section 70 can be configured as appropriate. In the example of FIG. 8, it includes a target designation section 71, candidate images 72, and a candidate arrangement line 73. The target designation section 71 is a part indicating the target (speculation target) for which the speculation information is to be displayed. The target designation section 71 may be configured as appropriate as long as it can allow the user to identify the target. In the example of FIG. 8, it is indicated by a dashed line surrounding the card object CO (the card object CO with the front side hidden) that is the speculation target. Also, the speculation information section 70 may be displayed so as to make an appropriate card object CO the speculation target. For example, it may be displayed so as to target all face-down card objects CP, or it may be displayed so as to target an appropriate part of them. In the example of FIG. 8, it is displayed so as to target the face-down card object CP designated (selected) by the user. In this case, the target designation section 71 has a function of distinguishing the card object CP designated as the speculation target by the user from other face-down card objects CP.

[0090] The candidate image 72 is an image showing candidates for the card object CO to be estimated. More specifically, an image of the front side (contents) of the card object CO estimated to correspond to the card object CO to be estimated is displayed as the candidate image 72. Only the card object CO with the highest probability as the candidate image 72 may be displayed, but in the example of FIG. 8, two candidate images 72 ("A" and "B") corresponding to the top two with high probabilities (an example and may be an appropriate number) are displayed. Further, a numerical value indicating the probability (hereinafter sometimes referred to as a matching probability value) is displayed above each candidate image 72. For example, a numerical value of "56%" is displayed above the candidate image 72 corresponding to "A", and a numerical value of "35%" is displayed above the candidate image 72 corresponding to "B". These matching probability values are output by a learned artificial intelligence model (estimation program FP) for the estimation service based on the situation log SL.

[0091] The candidate array line 73 is a line showing the relevance between the target specifying unit 71 and each candidate image 72. Each candidate image 72 is arranged below the candidate array line 73 along the line 73. Therefore, the candidate array line 73 also functions as a reference for arranging the candidate images 72 side by side horizontally.

[0092] (Calculation method of estimation information) Next, with reference to FIG. 9, a method for calculating inference information (candidates and match probability values) output by a learned artificial intelligence model for an inference service will be described. FIG. 9 is an explanatory diagram for explaining an example of a method by which an artificial intelligence model calculates inference information. The example of FIG. 9 shows a case where inference information is provided on the battle screen 50. The artificial intelligence model for the inference service may also be obtained by an appropriate learning method, but the example of FIG. 9 shows a case where it is generated by a neural network (deep learning) method. As shown in FIG. 9, in this case, similar to the calculation of probability values, inference information is calculated using an input layer, an intermediate layer, and an output layer. Specifically, in the input layer, information indicating the situation of the battle screen 50 is input as input information (input values) into the learned artificial intelligence model for the inference service. Although information different from the case of calculating probability values may be used as input information, as an example, a similar situation log SL (only a part is shown in FIG. 9) is used.

[0093] In the intermediate layer, an output value is calculated from the input value in the input layer using a predetermined functional formula that utilizes appropriate weighting (weight values). However, the weight values and the functional formula used therein are different from those of the artificial intelligence model for the analysis service. The weight values are defined by the inference learned data FD (data generated by the above learning), and the functional formula is configured to infer the content of the back-facing card object CO. Although an appropriate number of intermediate layers may be formed in the calculation of inference information, two intermediate layers are formed in the example of FIG. 9. Similarly, an appropriate number of output values (nodes) may be provided in each intermediate layer, but four output values are provided in each intermediate layer in the example of FIG. 9.

[0094] In the output layer, a value indicating the possibility corresponding to the card object CO to be inferred is calculated for each card object CO from the output value of the final intermediate layer (the second layer). The card objects CO for which the possibility is calculated may be all types that can be used in the card game (the opponent's deck cards may not be considered), but as an example, it is limited to the card objects CO incorporated in the opponent's deck cards. This limitation may be realized as appropriate. For example, when a learning dataset that includes only the opponent's deck cards as performance samples is prepared in advance and a learned artificial intelligence model (inference model unit 24) that has learned the learning dataset is provided, it may be limited by using the artificial intelligence model. As an example, it is realized by excluding (filtering) cards other than the opponent's deck cards from the candidates after calculating the inference candidates.

[0095] Also, in the card game, the rules are reflected as appropriate. For example, there may be a rule that limits the number of cards of a specific type of card object CO that can be incorporated into the deck to three. In this case, if all three cards of that specific type of card object CO have already been revealed (for example, if they are placed face-up in the field area 54, etc.), that specific type of card object CO is excluded (filtered) from the target of the inference information.

[0096] In addition, rule changes may occur as appropriate at a predetermined time (including regular times such as once a year, and temporary times for responding to situations at any time, etc., non-regular times). For example, the number of a specific type of card object CO that can be incorporated into the deck cards may change from three to two, etc., and the available number may change. Similarly, regulations may be changed as appropriate, such as limit regulations (restrictions that make a specific card object CO unusable). For example, if there is a card object CO that is too powerful, there may be a bias in the deck cards due to that card object CO always being incorporated, etc. Or, there may be a sense of unfairness in the battle results. Therefore, for example, when a specific card object CO is prohibited from being used, that specific card object CO is excluded (filtered) from the objects for which possibilities are calculated. Or, when the number of cards that can be used is changed to two, when both cards are open, that card object CO is excluded from the target of the speculation information (when the limit number increases, the exclusion from the target is postponed until it corresponds to the limit number).

[0097] In a card game, as described above, various changes such as changes in numbers or types can occur in the card objects CO (elements) that can be used according to rule changes, etc. And that change is reflected in the target of the speculation information. Specifically, the range in which a value indicating a possibility is calculated (in other words, the card objects CO that correspond to the candidates for the hidden card objects CO) is limited to the card objects CO that are usable in the current rules and that the opponent has incorporated into the deck cards.

[0098] Also, in the output layer, the values indicating possibilities (the values of each candidate after being limited to candidates available for use by the filter) are normalized such that the sum of the values of all options (e.g., all card object COs incorporated by the opponent into the deck cards) selectable by a normalization function (e.g., the Softmax function) becomes 100%. Then, the value after such normalization is calculated as inference information (match probability value). As an example, for each card object CO to be inferred (card object CO whose content is concealed), the candidate card object CO and the match probability value are calculated by a learned artificial intelligence model for the inference service by such a method. Then, the top (e.g., two in the example of FIG. 8) of the calculated match probability values among the candidate card object COs is provided to the user via the inference information unit 70. Also, when the information input in the input layer includes executable actions (options), even if the situation is the same, how it was used to reach the current situation (board state) is considered. Thereby, it is possible to realize the calculation of highly accurate inference information.

[0099] (Type of inference information) Next, with reference to FIGS. 10 to 11, the types of inference information that can be provided via the inference service will be described. The learned artificial intelligence model for the inference service may be only one type, but as an example, a plurality of types of artificial intelligence models (for example, the first inference program FP1, the second inference program FP2, etc.) are prepared. Each artificial intelligence model outputs inference information based on different learned inference data FD for inference. Since different artificial intelligence models may reach the same conclusion, that is, the same inference information, it is not always the case that different inference information is output by a plurality of types of artificial intelligence models, but in many cases, a plurality of types of inference information are output. FIG. 10 is an explanatory diagram for explaining an example of the types of inference information output by a plurality of types of artificial intelligence models. In addition, various inference information can be appropriately output by each artificial intelligence model, but in the example of FIG. 10, (1) recommended information for each period and (2) recommended information for each rank are shown. Furthermore, appropriate recommended information (including all of them) among these plurality of recommended information may be displayed as a display target, but in the example of FIG. 10, only a part of the recommended information selected by the user is displayed in both (1) and (2).

[0100] The recommended information for each period is the recommended information output by a learned artificial intelligence model based on the learned inference data FD for inference generated by the learning data set for each period. In this case, the plurality of learned inference data FD are respectively generated based on a plurality of learning data sets (each including performance samples for a plurality of periods). The inference information for each period may be provided as appropriate, but as an example, it is switched by a user's instruction (selection). The user's instruction may be executed as appropriate, but in the example of FIG. 10, the case of being executed via the information switching unit 80 is shown.

[0101] On the one hand, the recommended information for each rank is the recommended information output by a learned artificial intelligence model based on the learned inference learning data FD for each rank generated by the learning dataset for each rank. The card game is provided with a function of ranking players according to their achievements (information indicating the skill level of the players), and the recommended information for each rank corresponds to the output result of the artificial intelligence model whose learning dataset containing achievement samples of the same rank is the source. In this case, a plurality of inference learning data FDs corresponding to a plurality of learning datasets (each containing achievement samples of a plurality of ranks) are prepared. The inference information for each rank may be provided as appropriate, but as an example, it is provided so as to be switched according to the user's instruction via the information switching unit 80, similar to the inference information for each period. In the example of FIG. 10, the inference information unit 70 in the case where the battle screen 50 includes the information switching unit 80 is enlarged and schematically shown in both (1) and (2).

[0102] As shown in (1) and (2) of FIG. 10, the information switching unit 80 includes an information type unit 81, a first switching unit 82, and a second switching unit 83. The information type unit 81 is a part for displaying the type of the inference information that is currently the display target. Both the first switching unit 82 and the second switching unit 83 are parts indicating the positions where an instruction (a touch operation as an example) for switching the type of the inference information to be displayed should be executed. When a touch operation is executed on the first switching unit 82 and the second switching unit 83, the current display target is switched to the previous display target and the subsequent display target, respectively.

[0103] Specifically, as shown in (1) of FIG. 10, when the estimated information for each period is provided, the type of the estimated information is expressed by the information of "23.1 - 23.2" indicating the period as an example. The information of "23.1 - 23.2" (indicating a one - month period from January to February 2023) indicates the period during which the performance samples included in the generated learning dataset were collected. On the other hand, when a touch operation is executed on the first switching unit 82, the display target is switched to a period before the currently displayed target period (for example, a one - month period from December 2022 to January). That is, the display target is switched to the estimated information (candidate image 72 including a coincidence probability value) output by the artificial intelligence model generated by a plurality of learning datasets each including the performance samples of the previous period. Similarly, when a touch operation is executed on the second switching unit 83, the display target is switched to a period after the currently displayed target period (for example, a one - month period from February to March 2023). In the example of FIG. 10, the range of the display target is divided monthly, but it is not limited to this and may be appropriate as needed.

[0104] Also, as shown in (2) of FIG. 10, when the estimated information for each rank is provided, the type of the estimated information is expressed by the information of "second rank" indicating the rank. The information of "second rank" indicates the rank of the player during which the performance samples included in the generated learning dataset were collected. In this case, it means that a plurality of learning datasets each including the performance samples only of the players of the second rank are the generation sources. On the other hand, when a touch operation is executed on the first switching unit 82, the display target is switched to a rank lower than the currently displayed target rank (for example, the first rank). That is, the display target is switched to the estimated information (candidate image 72 including a coincidence probability value) output by the artificial intelligence model generated by a plurality of learning datasets each including the performance samples only of the players of the first rank, which is one rank lower than the current second rank. Similarly, when a touch operation is executed on the second switching unit 83, the display target is switched to a rank higher than the currently displayed target rank (for example, the third rank).

[0105] In the example of FIG. 10, the display target range is divided for each rank, but it is not limited to this, and it may be an appropriate range (for example, two or more ranks, etc.). In this case, information indicating the range may be added. Also, the initial display target (default display) may be set as appropriate, for example, to the user's own rank, or the rank of the opponent in the game, etc. In this case, since the speculation information adjusted to the user's own or the opponent's rank is displayed by default, the usability can be improved. Similarly, the initial display during a period may be set as appropriate, for example, to the latest period, or the period with the highest usage frequency, etc.

[0106] Each artificial intelligence model can output various types of speculation information based on differences such as algorithms. As described above, in the example of FIG. 10, multiple types of speculation information are output based on the differences in the learned data FD for speculation. In a card game, there may be a trend in the types of card objects CO used (combinations of deck cards). When there is a trend, the artificial intelligence model that reflects the trend is more likely to output more accurate speculation information. And trends often occur for each period. Therefore, when speculation information for each period is provided, the trend can be reflected in the speculation information, and thus more accurate speculation can be achieved. Similarly, in card objects, there are often differences in the card objects CO incorporated into the card deck and the usage methods, etc., according to the rank (the player's skill level). Therefore, when speculation information for each rank is provided, the usage tendency for each rank can be reflected in the speculation information, and thus more accurate speculation can be achieved.

[0107] Note that the learning dataset for each of the above-mentioned periods or for each rank is not limited thereto, and as a plurality of learning datasets, appropriate learning datasets including achievements collected from various viewpoints such as rules (regulations), the number of available cards, etc. may be used. For this reason, for example, a learning dataset including performance samples regarding all card objects CO (all not limited by period or rank) included in a card game may be used. That is, the plurality of inferred learned data FDs may be data that gives appropriate thinking to an inference program generated by various learning datasets.

[0108] FIG. 11 is an explanatory diagram for explaining a learning dataset for realizing inference information for each period and for each rank. The inference information for each period and for each rank may be prepared separately as independent inference information, and may be realized by separate learning datasets such as, for example, a plurality of learning datasets for each period or a plurality of learning datasets for each rank. On the other hand, inference information in which both are specifically considered, such as inference information for each period being prepared for each rank, may be prepared. The example of FIG. 11 shows the types of learning datasets when inference information in which both are considered is prepared.

[0109] As shown in FIG. 11, when inference information in which both are considered is prepared, learning datasets of types corresponding to the number of periods × the number of ranks are prepared. Specifically, the learning dataset corresponding to the first rank (including the performance samples of the players of the first rank) is prepared for each period such as the "first learning dataset" corresponding to the period of "January 2023 - February 2023" and the "second learning dataset" corresponding to the period of "January 2023 - February 2023" (both learning datasets include the performance samples of the players in the corresponding period). In this case, for example, the artificial intelligence model generated by the "first learning dataset" outputs inference information considering (reflecting) both the tendencies of the players of the first rank and the tendencies of the deck cards that were popular from January to February 2023.

[0110] The same applies to the second rank and the third rank, and learning datasets for each period such as the "third learning dataset" to the "sixth learning dataset" are prepared respectively. And the artificial intelligence models based on them output speculation information considering (reflecting) the trends of each rank and each period. In the example of FIG. 11, the first rank to the third rank are shown, but the card game may be provided with an appropriate number of ranks, and learning datasets corresponding to the number may be prepared. The same applies to the period.

[0111] (Processing of the network system) Next, referring to FIGS. 12 to 14, as an example of the processing of the network system 1, the option analysis processing, the display change processing, and the backside speculation processing will be described. The option analysis processing is a process for providing multiple types of superiority and inferiority information for each option of the selection opportunity by using multiple types of artificial intelligence models. The option analysis processing may be configured to provide multiple types of superiority and inferiority information by using multiple types of artificial intelligence models in parallel, but the example of FIG. 12 shows the case of providing multiple types of superiority and inferiority information by using multiple types of artificial intelligence models in series. Further, the option analysis processing is realized by the cooperation of the analysis model unit 23 of the game server 2 and the model processing unit 35 of the user device 3. In the example of FIG. 12, the processing mainly executed by the analysis model unit 23 is shown as the game server 2, and the processing mainly executed by the model processing unit 35 is shown as the user device 3, respectively.

[0112] When a predetermined analysis time (for example, the user's turn in the battle screen 50 or each phase) arrives, the model processing unit 35 starts the option analysis process of FIG. 12 and first requests the game server 2 to analyze each option in the selection opportunity (step S101). This request may be executed as appropriate. For example, it may be executed against the analysis model unit 23, and in the analysis model unit 23, the analysis results of each analysis model unit 23 such as the first analysis model unit 23A and the second analysis model unit 23B may be output in a predetermined order. However, as an example, it is executed against any one of the analysis model units 23 such as the first analysis model unit 23A and the second analysis model unit 23B in a predetermined order. Specifically, in step S101, the model processing unit 35 first requests the first analysis model unit 23A to analyze each option. Also, this request includes a situation log SL indicating the situation at the time of the request.

[0113] When the analysis request is sent, the first analysis model unit 23A starts the option analysis process of FIG. 12 and first acquires the request (step S201). Subsequently, the first analysis model unit 23A executes the analysis of each option based on the situation log SL included in the acquired request (step S202). This analysis may be executed only for some of the options in the selection opportunity, but as an example, it is executed for all options. After executing the analysis, the first analysis model unit 23A outputs the analysis result (step S203) and sends it to the model processing unit 35 (step S204). Then, after the transmission, the first analysis model unit 23A ends the current option analysis process.

[0114] On the other hand, when the model processing unit 35 receives the analysis result from the first analysis model unit 23A, it acquires the analysis result and determines whether the number of types of the acquired analysis result (in other words, the artificial intelligence model) has reached the set number (for example, two types: attack type AI and defense type AI) (step S102). If the number of types of the acquired analysis result has not reached the set number (step S102: No), the model processing unit 35 returns to step S101 and requests the next analysis model unit 23 (the second analysis model unit 23B) to perform analysis in a predetermined order (step S101).

[0115] When the analysis request is sent, the second analysis model unit 23B starts the option analysis process in FIG. 12, first acquires the request (step S201). Then, similar to the first analysis model unit 23A, it executes the analysis of each option (step S202), outputs the analysis result (step S203), and sends it to the model processing unit 35 (step S204). After the transmission, the second analysis model unit 23B ends the current option analysis process.

[0116] On the other hand, when the number of types of the acquired analysis result has reached the set number (step S102: Yes), the model processing unit 35 extracts the analysis result (superiority and inferiority information) to be displayed from all the acquired analysis results (including the analysis results of multiple types of analysis model units 23) (step S103). The analysis result to be displayed can be set appropriately. As an example, it is set to an analysis result showing the top five probability values across the analysis results by multiple types of analysis model units 23. Therefore, in step S103, the model processing unit 35 extracts the analysis results corresponding to the top five probability values among all the analysis results as the display targets. Subsequently, the model processing unit 35 displays an advice display column 60 including the five analysis results extracted in step S104 as the advice part 61 on the battle screen 50. After this display, the model processing unit 35 ends the current option analysis process. Thereby, the analysis service on the battle screen 50 is realized via the advice display column 60. More specifically, a battle screen 50 including an advice display column 60 that strictly selects and provides the top five pieces of information on the probability values output from multiple artificial intelligence models for each option is realized.

[0117] The display change process is a process for changing the display mode of each analysis result in the advice display column 60, such as the display target or the sorting order. The display mode of the advice display column 60 may be fixed, for example, as five in the order of the top probability values, but as an example, it is variable and changes according to the user's designation. When the change of the display mode of the advice display column 60 is instructed by the user, the model processing unit 35 starts the display change process of FIG. 13 and first discriminates the conditions designated by the user (step S301). Although arbitrary designations may be allowed for the user, as an example, it is limited to the conditions prepared in advance. Also, the conditions prepared in advance may include appropriate conditions, for example, include filter conditions and sorting conditions.

[0118] The filter condition is a condition for partially limiting the analysis result of the display target. The filter condition may be various conditions for appropriately limiting the display target, but for example, includes the case of limiting to the analysis result of a specific artificial intelligence model (when three or more types of analysis model units 23 are prepared, it may also be two or more types of analysis model units 23) based on the user's designation. The sorting condition (sorting condition) is a condition for sorting the analysis results in a predetermined order designated by the user. For this reason, the model processing unit 35 discriminates conditions such as filter conditions or sorting conditions as the designated conditions in step S301. In the case of filter conditions, the model processing unit 35 also discriminates the specific artificial intelligence model designated as the filtering target as the designated condition. On the other hand, in the case of sorting conditions, the model processing unit 35 discriminates the predetermined sorting order designated as the sorting condition as the designated condition. Both filter conditions and sorting conditions may be designated, and in that case, both are discriminated.

[0119] Subsequently, the model processing unit 35 determines the analysis result (superiority / inferiority information) of the target corresponding to the specified condition determined in step S301 from all the analysis results output by the analysis model unit 23 (all the analysis results obtained in step S102 of the option analysis process in FIG. 12) (step S302). Specifically, the model processing unit 35 determines the analysis result output by a specific artificial intelligence model (a type of the specific analysis model unit 23 such as the first analysis model unit 23A or the second analysis model unit 23B) specified as a filter condition among all the analysis results. Alternatively, the model processing unit 35 determines each analysis result corresponding to the order of a predetermined order. The predetermined order may allow appropriate designation, for example, the order for each artificial intelligence model such as the second analysis model unit 23B (defensive AI), the first analysis model unit 23A (offensive AI), the order from the lowest probability value, the top order within a predetermined range of probability values, or the order of the total value of probability values by a plurality of artificial intelligence models (the top order, the bottom order, etc. may be appropriate).

[0120] Next, the model processing unit 35 changes the display of the advice display column 60 so as to display the analysis result of the target determined in step S302 according to the specified conditions (step S303). For example, when the analysis result of the first analysis model unit 23A (for example, an attack-type AI) is specified as the filter condition, the display of the advice display column 60 is changed so that only each analysis result of the first analysis model unit 23A is displayed. When the sorting order for each artificial intelligence model is specified as the sorting condition, the display of the advice display column 60 is changed so that the advice parts 61 are arranged for each artificial intelligence model. The same applies when the sorting order such as in ascending order of the low probability value, in descending order of the probability value within a predetermined range, or in descending order of the total value of the probability values by a plurality of artificial intelligence models is specified. In these cases, the number of display targets may change as appropriate, but as an example, it is set to five, the same as before the change. For this reason, the model processing unit 35 changes the display of the advice display column 60 so as to display the advice part 61 corresponding to the top five probability values of the first analysis model unit 23A or the top five probability values in the specified sorting order. Then, after this change, the model processing unit 35 ends the current display change process. As a result, the display of the advice display column 60 is changed so as to display the analysis result specified by the user in the specified sorting order or the like.

[0121] The backside speculation process is a process for providing speculation information regarding the card object CO (information not shared with the user) facing the backside of the opponent in the battle screen 50 by using a learned artificial intelligence model. The backside speculation process is realized by the cooperation of the speculation model unit 24 of the game server 2 and the model processing unit 35 of the user device 3. In the example of FIG. 14, the process mainly executed by the speculation model unit 24 is shown as the game server 2, and the process mainly executed by the model processing unit 35 is shown as the user device 3, respectively.

[0122] When a predetermined start condition is satisfied (for example, when a card object CO facing the back side is selected by a touch operation or the like by the user on the battle screen 50), or when a change in the type of speculation information is instructed via the first switching unit 82 or the like of the information switching unit 80, the model processing unit 35 starts the back side speculation process of FIG. 14. First, it requests the game server 2 to speculate on the card object CO facing the back side of the target on the battle screen 50 (the card object CO selected by the touch operation or the card object CO being selected before the switching instruction) (step S401). This request includes a situation log SL indicating the situation at the time of the request. When a change in the type of speculation information is instructed via the first switching unit 82 or the like of the information switching unit 80, the request also includes information on the type of speculation information specified by the instruction.

[0123] When the speculation request is sent, the speculation model unit 24 starts the back side speculation process of FIG. 14 and first acquires the request (step S501). Subsequently, the speculation model unit 24 executes speculation on the content of the card object CO facing the back side based on the situation log SL included in the acquired request (step S502). When a change in the type of speculation information is instructed via the first switching unit 82 or the like of the information switching unit 80, this speculation is executed by a speculation model unit 24 of the type corresponding to the instructed type, such as the first speculation model unit 24A and the second speculation model unit 24B. Then, the speculation model unit 24 outputs the speculation result of the specified artificial intelligence model such as the first speculation model unit 24A (step S503). In this case, the speculation model unit 24 may output the speculation result only for the number (for example, the top two) to be displayed as the candidate image 72, but as an example, it outputs including the information of all candidates. Also, the speculation model unit 24 may output the speculation information regarding all the card objects CP facing the back side included in the battle screen 50 as the speculation result (in this case, when an object other than the current specified object is newly specified as the speculation target later, the user device 3 may extract and display the speculation information of the newly specified object from the already output speculation information), but as an example, it outputs only the speculation information regarding the card object CP facing the back side specified this time.

[0124] Next, the inference model unit 24 adjusts the inference result (step S504). The adjustment can be realized as appropriate. As an example, it is executed to reflect rules such as limit regulation and the opponent's deck cards. For example, the inference model unit 24 adjusts the inference result so as to exclude the card object CO that has been prohibited from use and the card object CO that the opponent has not incorporated into the deck cards from the output inference result. Also, this exclusion may be executed by simply excluding the card object CO such as prohibited from use from the candidates output as the inference result, but as an example, it is executed so that the coincidence probability value is recalculated (adjusted) after the exclusion. After that, the inference model unit 24 transmits the adjusted inference result to the model processing unit 35 as inference information (step S505). Then, after the transmission, the inference model unit 24 ends the current reverse-side inference process. Note that when the process is executed by an artificial intelligence model that has learned the learning data set limited to the adjusted target as described above (when the same result can be obtained without adjustment), etc., in the case where adjustment is not required (including the case where adjustment is not executed in the first place), the process of step S504 may be omitted as appropriate.

[0125] On the other hand, when the inference information is transmitted from the inference model unit 24, the model processing unit 35 acquires the inference information (step S402). Subsequently, the model processing unit 35 specifies the inference information to be displayed from the acquired inference information (step S403). For example, when the number to be displayed as the candidate image 72 in the inference information unit 70 is the top two in terms of the coincidence probability value, the inference information corresponding to the top two in terms of the coincidence probability among the acquired inference information is specified as the display target. Next, the model processing unit 35 displays the inference information of the display target specified in step S404 on the battle screen 50 (step S404). Specifically, the inference information unit 70 including the inference information specified in step S404 is displayed on the battle screen 50. Then, after this display, the model processing unit 35 ends the current reverse-side inference process.

[0126] According to the procedure of FIG. 14, the speculation service on the battle screen 50 is realized via the speculation information unit 70. More specifically, as information for speculating on the content of the card object CO facing the back side, speculation information (for example, information including the content of the candidate card object CO and the coincidence probability) output from the artificial intelligence model is strictly selected and provided for the top two for each card object CO facing the back side. A battle screen 50 including a speculation information unit 70 is realized. Note that the display target or the order of the speculation information unit 70 may be changed as appropriate. In that case, for example, the display target and the like may be changed according to the same procedure as the display change process in the example of FIG. 13 (however, the procedure targeting speculation information and the speculation information unit 70 instead of the analysis result and the advice display column 60).

[0127] As described above, according to this form, via the analysis service, a plurality of probability values (superiority and inferiority information) respectively indicating the probability of winning the battle for each option on the battle screen 50 by a plurality of types of learned artificial intelligence models (the first analysis model unit 23A, the second analysis model unit 23B) (which may be information indicating various possibilities of proceeding advantageously within a predetermined range of the analysis target) are output, and information on the top five of these probability values is provided before the selection of the option. Therefore, in order to win the battle, it is possible to provide numerical information indicating the magnitude of the influence on winning in the game, that is, information on the probability values (superiority and inferiority information) from various viewpoints for each selection option such as which card object CO should be placed in which card placement area CP or whether the phase should be ended.

[0128] In a card game, turns including a plurality of phases are alternately provided to players, and a selection opportunity is provided for each phase. However, the end of a phase or the end of a turn is also included in the group of options in each selection opportunity. When the alternation of turns or the like functions as one of the options, the influence of the selection in the selection opportunity on the progress tends to become more complicated. When a plurality of selection opportunities are included in each turn through a plurality of phases and each of these selection opportunities ends selectively, it is highly likely to become even more complicated. Originally, since a complicated selection including many options (including options further derived from each option) is required in each selection opportunity, even if the turn instruction or the like is not selective, a complicated selection is still required. Regarding such a complicated selection, information on a plurality of probability values can be provided from various viewpoints. Furthermore, the information on the plurality of probability values is provided with some being limited through a filter condition or the like and being appropriately rearranged. Therefore, the convenience of accessing information on a specific probability value (option information) through a filter condition or the like can be improved.

[0129] On the other hand, through the inference service, information on the card object CO (element) of the opponent that is not shared with the user, for example, the content of the card object CO hidden from the user by being placed face down, is inferred by a learned artificial intelligence model (inference model unit 24), and the inference result is provided as inference information. Therefore, the inference information can assist in inferring the card object CO hidden in the incomplete information game.

[0130] Also, in a card game, among all card objects CO, the deck cards selected by the user from the owned cards are used by the user. Therefore, the card game is configured as a competitive game of a type where there are differences in assets between players. In this case, it is highly likely that it is more difficult to infer the card objects CO that are hidden compared to the case where all element groups are used. Similarly, in a card game, the card objects CO that can be used may change depending on the time or the like, and it is also highly likely that it is more difficult to infer the card objects CO that are hidden compared to the case where they are fixed (do not change). For this reason, an experienced player may be able to infer the content of the card objects CO that are hidden, such as by looking at the back side, based on rules of thumb, but generally such inference is difficult, and it is often particularly difficult for beginners. Even when it is difficult to infer the hidden card objects CO like this, the inference can be assisted by inference information. Therefore, it is possible to help bridge the gap with experts (eliminate the gap) by the inference information. And when the change in the usable card objects CO is reflected in the inference information, the accuracy of the inference information can be improved compared to the case where the same change is not reflected. Therefore, the inference information can further help eliminate the gap with experts.

[0131] Furthermore, a plurality of pieces of inference information are provided by a plurality of learned artificial intelligence models according to the time period and rank. In card games, there may be trends in the types (element groups) of card objects used. When there is a trend, the artificial intelligence model that reflects the trend is more likely to output more accurate inference information. And trends often occur for each period. Similarly, there are often differences according to the rank (the skill level of the user) in the types and usage methods of the card objects CO incorporated in the group of card objects used. For this reason, by training a learning dataset including performance for each predetermined period, the trend for each period can be reflected in the inference information. Alternatively, by training a learning dataset including performance for each rank, the usage tendency for each rank can be reflected in the inference information. By these means, more accurate inference can be realized. Furthermore, the types of inference information (the target period and rank) can be switched according to the user's selection. For this reason, it is possible to select the type of inference information according to the opponent (whether the opponent is a beginner or which period's trend the opponent's deck cards correspond to, etc.) (in other words, the user's inference about the opponent can be reflected in the inference information), so the accuracy of the inference information can be further improved.

[0132] In the above form, the first analysis model unit 23A and the second analysis model unit 23B of the game server 2 (or the processes in FIG. 12 executed by them respectively) function as a plurality of types of processes of the present invention. On the other hand, the model processing unit 35 of the user device 3 functions as the superiority and inferiority acquisition means and the superiority and inferiority provision means of the present invention by executing the procedure in FIG. 12. Specifically, the model processing unit 35 functions as the superiority and inferiority acquisition means by executing step S102 in FIG. 12 and functions as the superiority and inferiority provision means by executing step S104. Also, the model processing unit 35 of the user device 3 functions as the filtering means or the sorting means of the present invention by executing the procedure of step S303 in FIG. 13.

[0133] In addition, the estimation model unit 24 of the game server 2 (or the process of FIG. 14 executed by the unit) functions as a predetermined process of the present invention. On the other hand, the model processing unit 35 of the user device 3 functions as the information acquisition means and the information providing means of the present invention by executing the procedure of FIG. 14. Specifically, the model processing unit 35 functions as the information acquisition means by executing step S402 of FIG. 14 and functions as the information providing means by executing step S404.

[0134] The present invention is not limited to the above-described form, and may be implemented in a form with appropriate modifications or changes. Further, the present invention may be implemented in a form obtained by appropriately combining various technical means included in the above-described form and forms with the following modifications and the like. In the above-described form, in a card game provided as a video game, superiority / inferiority information and estimation information are provided to the player. However, the present invention is not limited to such a form. For example, the superiority / inferiority information and the estimation information may be provided to the player or viewers watching the game in a game in the real space. Specifically, for example, in a game in the real space, the superiority / inferiority information or the estimation information may be provided by being added to a video (viewing video) of the game, or may be provided through sound to the real space where the game is played (including players and viewers).

[0135] In the above-described form, each analysis model unit 23 such as the first analysis model unit 23A (a learned artificial intelligence model for analysis services) and each estimation model unit 24 such as the first estimation model unit 24A (a learned artificial intelligence model for estimation services) are given different thoughts due to the difference in learned data such as the learned data AD for analysis or the learned data FD for estimation. However, the present invention is not limited to such a form. Each analysis model unit 23 and each estimation model unit 24 may be given different thoughts, for example, due to the difference in the algorithms of the learning program (rewards in reinforcement learning such as Q-learning, that is, including predetermined results) and the algorithms of the inference program.

[0136] In the above form, the superiority / inferiority information is provided via the advice display column 60. However, the present invention is not limited to such a form. The superiority / inferiority information may be provided as appropriate. FIG. 15 is an explanatory diagram for explaining an example of a modification regarding the display mode of the superiority / inferiority information. In the example of FIG. 15, two modifications are shown by (1) a first modification example and (2) a second modification example. As shown in (1) of FIG. 15, in the first modification example, the probability values of each analysis model unit 23 are provided via the target card image 90, the first analysis result unit 91, and the second analysis result unit 92.

[0137] The target card image 90 is an image showing the candidate card object CO as an option. The target card image 90 may be the card object CO itself or a thumbnail-like image replacing it. The first analysis result unit 91 and the second analysis result unit 92 are parts showing the types (types of analysis results) of the analysis model units 23, such as the first analysis model unit 23A (for example, an attack-type AI) or the second analysis model unit 23B (for example, a defense-type AI). The first analysis result unit 91 and the second analysis result unit 92 can be formed as appropriate, but in the example of FIG. 15, both are formed in a rectangle surrounding the target card image 90 and are shown by a dashed line and a thick solid line, respectively.

[0138] In addition, the first analysis result section 91 and the second analysis result section 92 are provided with a numerical information section 93 that shows the probability value (analysis result) by each analysis model section 23 such as "46.2". Furthermore, a total information section 94 may be provided. For example, in the first modification example of FIG. 15, three candidates are shown, and a total information section 94 is added to the middle candidate among them. The total information section 94 is a part that shows the total of the probability values of a plurality of types of artificial intelligence models for one candidate (option). The target card image 90 may be displayed for all candidates, but in the example of FIG. 15, it is displayed limited to a part such as the top three of the probability values. Since the probability values by a plurality of types of analysis model sections 23 are output for all three candidates, the total information section 94 may be displayed for all of them, but in the example of FIG. 15, it is displayed limited to a part (the middle target card image 90) of them. The display target of the total information section 94 can be set appropriately, but as an example, it is set to a candidate whose total value exceeds the highest probability value. For example, in the middle target card image 90, a numerical information section 93 showing "25.2" and "32.0" is provided in the first analysis result section 91 and the second analysis result section 92 respectively, and "57.2" showing the total value of them is shown as the total information section 94. And this total value is higher than any probability value of the other target card images 90. In such a case, the total information section 94 is provided.

[0139] Also, as shown in (2) of FIG. 15, in the second modification example, the probability values of the respective analysis model sections 23 are provided via the target card image 90, the first gauge section 95, and the second gauge section 96. The target card image 90 is the same as in the first modification example. The first gauge section 95 and the second gauge section 96 are parts that show each analysis model section 23 (type of analysis result) such as the first analysis model section 23A, and are configured to visually show the magnitude of the probability value. The first gauge section 95 and the second gauge section 96 can be formed appropriately, but in the example of FIG. 15, both are formed in a bar shape extending upward from a predetermined position according to the magnitude of the probability value, and are shown in white and black color schemes respectively.

[0140] In addition, a detailed information section 97 is provided in both the first gauge section 95 and the second gauge section 96. The detailed information section 97 may contain various types of information. In the example of FIG. 15, it includes information on the type of the analysis model section 23 (such as "attack-type AI" or "defense-type AI") like the first analysis model section 23A (for example, attack-type AI) or the second analysis model section 23B (for example, defense-type AI), and information on probability values ("46.1" or "32.0"). Although omitted in the example of FIG. 15, a totalization information section 94 may be provided as appropriate, and information on the total value may be displayed. The superiority / inferiority information may be provided as appropriate, and it is not limited to the example of FIG. 5. For example, it may be provided as in the modified example of FIG. 15.

[0141] In the above-described form, the network system 1 includes the game server 2. However, the present invention is not limited to such a form. For example, when an offline-type game that is played without connecting to the network NT is provided, etc., the game server 2 may be omitted, and the user device 3 may function alone as the game system of the present invention. In other words, various artificial intelligence models such as each analysis model section 23 may be provided in the user device 3. Therefore, for example, in the example of FIG. 12 or the example of FIG. 14, the processing (role) of the game server 2 may be executed by an artificial intelligence model (a logical device similar to the analysis model section 23, etc.) provided in the user device 3, which is called by the model processing section 35.

[0142] Alternatively, conversely, the game server 2 may execute all or part of the role (various processes, etc.) of the user device 3. For example, the learned data such as the learned data AD for analysis or the learned data FD for speculation may be stored in the game server 2, and the programs for the artificial intelligence model such as the analysis program AP or the speculation program FP may be provided in the user device 3. Also, data necessary for realizing the present invention, such as part by part of various learned data or part by part of the programs for the artificial intelligence model, may be appropriately and dispersedly recorded in the game server 2 and the user device 3. And the game system of the present invention may be realized by the cooperation between the game server 2 and the user device 3. The same applies to the game program PG2 and the game data GD. In these cases, combinations of the user device 3 and the game server 2 (for example, the network system 1), or the game server 2 alone (including the case where it is composed of a plurality of server devices), etc. may appropriately function as the game system of the present invention. And the programs and control methods implemented in devices such as the network system 1, the user device 3, or the game server 2 may function as the game program and control method of the present invention.

[0143] The various aspects of the present invention derived from each of the above-described embodiments and modification examples are described below. In the following description, corresponding members illustrated in the accompanying drawings are appended in parentheses in order to facilitate the understanding of each aspect of the present invention, but the present invention is not limited to the illustrated forms thereby.

[0144] The computer program of the present invention configures a computer (31) incorporated in a game system (3) that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence exerted by each option in a game including an opportunity to select a part of a plurality of options, for at least one of the plurality of options before the selection of the part of the options, to function as a superiority / inferiority acquisition means (35) that acquires a plurality of superiority / inferiority information for each option of the at least one option by using a plurality of types of processes (23A, 23B) that output the superiority / inferiority information for each option, and a superiority / inferiority providing means (35) that provides the plurality of superiority / inferiority information for each option before the selection of the part of the options.

[0145] According to the present invention, a plurality of superiority / inferiority information is output for each option by a plurality of types of processes, and they are provided as superiority / inferiority information before the selection of a part of the plurality of options. For this reason, in a game including an opportunity to select a part of a plurality of options, superiority / inferiority information can be provided from various viewpoints for each option. Note that the game of the present invention may be a game in the real space and does not necessarily need to be provided by a game system. Similarly, the plurality of processes do not necessarily need to be provided by a game system. For example, all or an appropriate part of the plurality of processes may be executed by a system different from the game system (the operator is also different).

[0146] In an appropriate game including a selection opportunity, superiority / inferiority information may be provided. For example, in a cooperative game including a selection opportunity (including the case of playing in a team), or a competitive game (including one-on-one, one-on-many (including the case where a plurality of users other than oneself each become an independent opponent in a game such as mahjong), and many-on-many), superiority / inferiority information may be provided. Similarly, superiority / inferiority information may be provided in a game simply played alone. Also, in each game, various effects given by each option may be output as superiority / inferiority information. For example, in a game with a concept of victory, the effect on victory may be output as superiority / inferiority information. Or, the effect on a part of the progress, such as a specific development provided in the game (including obtaining a specific mission and bonus scores, etc.), may be output as superiority / inferiority information. Specifically, for example, in one aspect of the computer program of the present invention, as the game, a competitive game in which the user plays against an opponent is provided, and the plurality of types of processes may be configured to output, as the plurality of superiority / inferiority information, any information regarding the superiority or inferiority of the effect given by each option on victory in the competitive game. In this case, information regarding the effect on victory in the competitive game can be provided from various viewpoints for each option.

[0147] The superiority / inferiority information may be appropriate information regarding superiority and inferiority. For example, it may be information of symbols for distinguishing the degree of superiority and inferiority such as A, B, and C, or information that sensually expresses superiority and inferiority such as "seems good" and "seems bad". Or, the superiority / inferiority information may be information in which superiority and inferiority are quantified. When superiority and inferiority are quantified, the numerical value may be a numerical value indicating the possibility of progressing advantageously (being less likely to become disadvantageous) within a predetermined range (including an appropriate range such as a predetermined number of selection opportunities, a predetermined time, or a predetermined development), or of course, a numerical value indicating the possibility of ultimately winning in a confrontation. For example, in an aspect where a competitive game is provided, the plurality of types of processes may be configured to output, as the plurality of superiority / inferiority information, any numerical information indicating the possibility of advantageous progress in the competitive game through the magnitude of the numerical value.

[0148] The competitive game may be configured as appropriate. For example, the competitive game may be configured such that the user and the opponent execute selections at any time (in parallel), or may be configured to give turns including alternating selection opportunities. Even when turns are given alternately, for example, a selection by a person outside the turn (this selection may function as the selection opportunity of the present invention) may not be completely excluded, such as when a selection by the opponent is required following the user's selection in the user's turn. Also, when turns are given alternately, the turn change may be automatically executed according to time, the number of instructions, etc., or may be executed as one of the selections in each turn. When the turn changes by a selection, the selection may function as the target of the superiority / inferiority information or may be excluded from the target of the superiority / inferiority information. Also, each turn may be configured as appropriate, for example, it may include only one selection opportunity or may include a plurality of selection opportunities. The plurality of selection opportunities may be appropriately distinguished through phases, etc., and the options in each selection opportunity may all be the same, or at least some (including all) of the options may be different. Also, the end of each phase may be automatic or selective. When the end of each phase is selective, the selection may or may not be the target of the superiority / inferiority information.

[0149] For example, in a mode where a battle game is provided, the battle game is configured such that turns including the selection opportunity are provided alternately to the user and the opponent, and the plurality of options may include an option corresponding to a turn end instruction for ending the user's turn. Also, in this mode, each turn includes a plurality of phases that each provide a plurality of selection opportunities, and the selection opportunities for each phase are configured to each provide, as the plurality of options, two or more options including different options among the selection opportunities for each phase, and the two or more options for each selection opportunity may each include an option corresponding to a phase end instruction for ending each phase. When the alternation of turns functions as one of the options, the influence of the selection in the selection opportunity on the progress tends to become more complicated. When there are a plurality of selection opportunities in each turn and each of those selection opportunities ends selectively, it is more likely to become even more complicated. When an option corresponding to a turn end instruction or the like is the subject of superiority / inferiority information, a plurality of superiority / inferiority information can be provided from various viewpoints regarding the complicated selection.

[0150] A game including a selection opportunity may be a game in the real space or an electronic game. Therefore, the superiority / inferiority information may be provided as appropriate. For example, in a game in the real space, the superiority / inferiority information may be provided by being added to an image (display on a monitor, display on a wearable terminal, etc.) that captures the state of the game, or may be provided through sound in the real space where the game is played. Similarly, in an electronic game, the superiority / inferiority information may be displayed on the game screen for the game or may be provided through sound. Also, the electronic game may be configured as appropriate, for example, configured to be provided via an appropriate object of a display device that displays a game screen including various objects. Specifically, in a mode where a battle game is provided, the battle game may be provided as a game that battles an opponent according to a predetermined rule via the plurality of objects of a display device (MO) that displays a plurality of objects (CO).

[0151] When a competitive game is provided via a plurality of objects, the plurality of objects may be used as appropriate. For example, when an electronic shogi or a chess game is provided, electronic pieces may be used as the plurality of objects. Alternatively, when an electronic mahjong or a card game is provided, electronic mahjong tiles or cards may be used as the plurality of objects. Also, in a game including various characters (including not only human-like entities but also various objects such as vehicles and animals), the plurality of characters may function as the plurality of objects. These objects may be used as appropriate according to the rules of the game. Similarly, various elements set along the rules of the game may function as a plurality of options. For example, each object (e.g., a character) may or may not function as an option.

[0152] Specifically, for example, in an aspect where a competitive game is provided, the competitive game is provided as a card game that uses a plurality of card objects (CO) as the plurality of objects, and the card game is played using a plurality of card placement areas (CP) where the card objects to be used in the game against the opponent are to be arranged respectively, a hand area (57) where a hand (CO5) as a candidate card object to be arranged in each card placement area is to be arranged, and a deck area (55) where a deck (CO6) as a candidate card object to be added to the hand is to be arranged, and the plurality of options may include options corresponding to at least one of the plurality of card placement areas, the hand to be arranged in each card placement area, and the addition of the deck to the hand. In this case, it is possible to provide a plurality of superiority and inferiority information from various viewpoints in a game that requires a complex selection including many options (including options further derived from each option).

[0153] The plurality of types of processing may be configured as appropriate as long as they can output a plurality of superiority / inferiority information. For example, each processing may be configured to output superiority / inferiority information according to a predetermined logic (including a predetermined rule, calculation formula, etc.). Alternatively, at least one of the plurality of types of processing may be configured as an artificial intelligence model (so-called AI) generated to output superiority / inferiority information. Specifically, in one aspect of the computer program of the present invention, at least one of the plurality of types of processing is configured as a learned artificial intelligence model generated by causing a predetermined pre-learning model to machine-learn a predetermined learning dataset so as to output information on superiority / inferiority regarding a predetermined result in the game as the superiority / inferiority information. Also, in this aspect, the plurality of types of processing may all be configured as the learned artificial intelligence model, and may be configured to output the plurality of superiority / inferiority information due to at least one difference among the predetermined result, the learning dataset, and the algorithm of the pre-learning model for learning the learning dataset.

[0154] The plurality of superiority / inferiority information may be provided as appropriate. For example, the superiority / inferiority information may be provided according to a pre-set order for each option, or when digitized, may be provided in an order according to the magnitude of the numerical value, such as in descending order. Also, the order of providing each superiority / inferiority information may be fixed or variable. For example, the plurality of superiority / inferiority information may be provided in the order specified by the user. Alternatively, the plurality of superiority / inferiority information may be provided with some being limited according to a predetermined filter condition. The filter condition may be fixed or variable. For example, the filter condition may be specified by the user. Specifically, in one aspect of the computer program of the present invention, the superiority / inferiority providing means may be provided with at least one of a filter means (35) for limiting a part of the plurality of superiority / inferiority information to satisfy a predetermined filter condition and a sorting means (35) for sorting the plurality of superiority / inferiority information according to a predetermined sorting condition. In this case, the plurality of superiority / inferiority information is provided with some being limited or appropriately sorted through a filter condition or the like. Therefore, the convenience of accessing specific superiority / inferiority information through a filter condition or the like can be improved.

[0155] On the other hand, the game system of the present invention is a game system (3) incorporating a computer (31) that provides a user with superiority-inferiority information regarding the superiority or inferiority of the influence exerted by each option in a game including a selection opportunity for selecting a part of a plurality of options, before the selection of the part of the options, for at least one of the plurality of options. The computer functions as a superiority-inferiority acquisition means (35) that acquires a plurality of pieces of superiority-inferiority information for each option of the at least one option by using a plurality of types of processes that output the superiority-inferiority information for each of the at least one option, and a superiority-inferiority providing means (35) that provides the plurality of pieces of superiority-inferiority information for each option before the selection of the part of the options.

[0156] Further, the control method of the present invention causes a computer (31) incorporated in a game system (3) that provides a user with superiority-inferiority information regarding the superiority or inferiority of the influence exerted by each option in a game including a selection opportunity for selecting a part of a plurality of options, before the selection of the part of the options, for at least one of the plurality of options, to execute a procedure of acquiring a plurality of pieces of superiority-inferiority information for each option of the at least one option by using a plurality of types of processes that output the superiority-inferiority information for each of the at least one option, and a procedure of providing the plurality of pieces of superiority-inferiority information for each option before the selection of the part of the options.

[0157] (Reference Example) In Non-Patent Document 1, a technique is disclosed in which information such as "winning rate", "candidate moves", and "lines of play" simulated by AI (artificial intelligence) for each move is added to a relay screen obtained by photographing a shogi board from above during a game. Games such as shogi are of a type called a so-called perfect information game in which all information about the pieces used by each player is disclosed. On the other hand, there also exist games of a type called imperfect information games. In that type of game, the game progresses without at least one element of a group of elements (for example, options) used by other users (including computers) being shared with one user. For this reason, in imperfect information games, information regarding elements that are not shared (hidden) with users may be important.

[0158] Therefore, the reference example aims to provide a computer program or the like that can assist in inferring elements hidden in an imperfect information game.

[0159] The computer program of the reference example configures a computer (31) incorporated in a game system (3) that provides element information as information regarding at least one of the element groups (CO) for each user in a game played by a plurality of users using the element groups for each user, to function as an information acquisition means (35) that acquires the inference information using a predetermined process (24) set so as to obtain inference information for inferring the content of the at least one element in an imperfect information game of a type in which at least one content of the element group of another user provided as the game is not shared with the one user, and an information provision means (35) that provides the inference information as the element information.

[0160] According to the reference example, the content of an element that is not shared with one user, that is, an element hidden from one user, is inferred by a predetermined process, and the inference result is provided as inference information. For this reason, the inference information can assist in inferring elements hidden in an imperfect information game. Note that the game of the reference example may be a game in the real space and does not necessarily need to be provided by a game system.

[0161] As an element group, various elements included in an incomplete information game may be appropriately used. For example, in the case where an incomplete information game such as mahjong or a card game is provided, elements such as the hands of other users or the cards in hand (the cards in one's possession) may be used as the element group. Also, like the mahjong tiles in mahjong, the element group for each user may be appropriately distributed and used from an element group (mahjong tiles) that is fixed for each user and specified in common for all users. The same applies to card games such as Seven Piles or poker. Alternatively, the incomplete information game may be played such that a partial element group arbitrarily designated by the user is used from the element groups associated with each user. For example, in one aspect of the computer program of the reference example, the element group of each user may be formed by two or more elements selected by each user from the denominator element group associated with each user as the denominator of the element group.

[0162] The incomplete information game may be configured as appropriate. For example, in the incomplete information game, the available elements may be fixed, or may be variable according to appropriate conditions such as time. Also, changes in the available elements may occur as appropriate. For example, the number or type of the available elements may change. That is, elements that become unavailable or newly added elements may occur according to a predetermined time or the like. When the available elements change, the change may or may not be reflected in the speculation information. Specifically, for example, in one aspect of the computer program of the reference example, the incomplete information game may be provided such that the available elements change at a predetermined time. Also, in this aspect, the change in the available elements may include at least one of a change in number and a change in type. Similarly, in the aspect where the available elements change, the predetermined process may be set to reflect the change in the available elements in the speculation information. When the available elements change according to time or the like, it is highly likely that it becomes more difficult to speculate on the elements to be concealed compared to the case where the available elements are fixed. Even when it is difficult to speculate on such concealed elements, the speculation can be assisted by the speculation information. And when the change in the available elements is reflected in the speculation information, the accuracy of the speculation information can be improved compared to the case where the same change is not reflected.

[0163] The incomplete information game may be configured as an appropriate game. For example, it may be configured as a cooperative game (including the case where it is played in a team), or a competitive game (including one-on-one, one-on-many (such as in mahjong where multiple users other than oneself become independent opponents), and many-on-many). Similarly, it may be configured as a game simply played by one person. Also, the incomplete information game may be a game in the real space or an electronic game. Therefore, the speculative information may be provided as appropriate. For example, in a game in the real space, the speculative information may be provided by being added to a video (such as display on a monitor or display on a wearable terminal) that captures the state of the game, or may be provided through sound in the real space where the game is played. Similarly, in an electronic game, the speculative information may be displayed on the game screen for the game or may be provided through sound. Also, the electronic game may be configured as appropriate. For example, it may be configured to be provided via an appropriate object of a display device that displays a game screen including various objects. Specifically, in one aspect of the computer program of the reference example, the incomplete information game may be provided as a card game in which the one user and the other user play against each other according to a predetermined rule via the plurality of card objects (CO) of a display device (MO) that displays a plurality of card objects functioning as an element group of each user. Also, in this aspect, the display device may display a game screen (50) including the plurality of card objects, and the game system may provide the card game as a video game via the game screen.

[0164] Also, when a card game is provided as an incomplete information game, changes in available elements (e.g., whether an element can be used due to rule changes, etc.) may or may not be reflected in the speculation information as described above. If they are reflected in the speculation information, such reflection may be realized as appropriate. For example, after a speculation result is obtained, the unavailability may be reflected in the speculation information simply by excluding the elements that cannot be used from the speculation result. Alternatively, after the elements that cannot be used are excluded, the speculation result may be adjusted again (e.g., reflecting the influence of the unavailability in other speculations) so that the unavailability is reflected in the speculation information. In any case, the unavailability may be reflected such that the card object that cannot be used is excluded from the speculation information. Specifically, for example, in an aspect where a card game is provided as an incomplete information game, changes in the card objects that can be used in the card game include changes that make specific card objects unavailable due to the change of the rules, and the predetermined process may be set to reflect the unavailability of the specific card object in the speculation information by excluding the specific card object.

[0165] The card game may be played according to appropriate rules. For example, the card game may be played using fixed cards. Alternatively, the card game may be configured to grant appropriate card objects to users who meet predetermined granting conditions, and some of the card objects are owned by the users, and the card game may be played using the card objects owned by each user. All of the card objects owned by each user may be used in the play, or only an appropriate part, such as the card objects selected by each user, may be used in the play. Specifically, for example, in the aspect where a card game is provided as an incomplete information game, the card game is configured to grant a card object (CO) to a user who meets a predetermined granting condition, and the denominator element group may be composed of the card objects (CO) granted to each user. In this case, compared with the case where the available elements are fixed, the candidates for the elements to be hidden change for each play, so the speculation is likely to be more difficult. And even when it is difficult to speculate on such hidden elements, the speculation can be assisted by the speculation information.

[0166] When a card game is provided as an incomplete information game, the card game may be configured to appropriately use a plurality of card objects (element groups). And an appropriate part of the plurality of card objects may correspond to the object of speculation information. For example, in the aspect where a card game is provided as an incomplete information game, the card game is played using a plurality of card placement areas (CP) where the card objects to be used in the battle with the other users should be arranged respectively, a hand area (57) where a hand (CO5) as a candidate card object to be arranged in each card placement area should be arranged, and a deck area (55) where a deck (CO6) as a candidate card object to be added to the hand should be arranged, and the predetermined process may be set so that the content of at least one card object of the plurality of card placement areas, the hand, and the deck among the plurality of card objects of the other user can be obtained as the speculation information.

[0167] The predetermined process may be appropriately configured as long as speculative information can be obtained. For example, each process may be configured to output speculative information according to a predetermined logic (including predetermined rules, calculation formulas, etc.). Alternatively, the predetermined process may be configured as an artificial intelligence model (so-called AI) generated so as to obtain speculative information. Further, the predetermined process may be configured as one type of process for obtaining one type of speculative information, or may be configured as a plurality of processes for obtaining a plurality of pieces of speculative information regarding one confidential element. For example, in one aspect of the computer program of the reference example, the predetermined process may be configured as a learned artificial intelligence model generated so as to obtain the speculative information by machine learning a predetermined learning dataset. Further, in this aspect, the learned artificial intelligence model may include a plurality of learned artificial intelligence models (24A, 24B) respectively generated by machine learning a plurality of learning datasets (FD) having different contents. In a card game, there may be a trend in the types (element groups) of card objects used. When there is a trend, an artificial intelligence model reflecting the trend is more likely to output more accurate speculative information. And trends often occur for each period. Similarly, there are often differences in the types and usage methods of card objects incorporated in the card object group used according to the rank (the skill level of the user). For this reason, by learning a learning dataset including performance for each predetermined period, the trend for each period can be reflected in the speculative information. Alternatively, by learning a learning dataset including performance for each rank, the usage tendency for each rank can be reflected in the speculative information. By these means, more accurate speculation can be realized.

[0168] When the learned artificial intelligence model includes a plurality of learned artificial intelligence models, a plurality of pieces of inference information obtained by the plurality of learned artificial intelligence models may be provided, or a part of the plurality of pieces of inference information may be provided. When a part of the plurality of pieces of inference information is provided, the part may be specified as appropriate. For example, it may be specified according to the play situation, or may be specified according to an evaluation (hit rate, evaluation by the user, etc.) over a certain period. Alternatively, it may be specified based on a designation by the user. For example, in an aspect where the learned artificial intelligence model includes a plurality of learned artificial intelligence models, the inference information includes a plurality of pieces of inference information respectively obtained using the plurality of learned artificial intelligence models, and the information providing means may provide the inference information corresponding to the learned artificial intelligence model selected by the one user among the plurality of pieces of inference information to the one user.

[0169] On the other hand, the game system of the reference example is a game system incorporating a computer that provides element information as information regarding at least one of the element groups (CO) for each user in a game played by a plurality of users using the element groups for each user. The computer uses a predetermined process (24) set so as to obtain inference information for inferring the content of the at least one element in an incomplete information game of a type in which the content of at least one of the element groups of other users provided as the game is not shared with one user, and functions as information acquisition means (35) for acquiring the inference information and information providing means (35) for providing the inference information as the element information.

[0170] Also, in a game played by a plurality of users using an element group (CO) for each user, in a computer (31) incorporated in a game system (3) that provides element information as information regarding at least one of the element groups for each user, in an incomplete information game of a type in which the content of at least one of the element groups of other users provided as the game is not shared with one user, a procedure for acquiring the speculation information using a predetermined process (24) set so as to obtain speculation information for speculating the content of the at least one element, and a procedure for providing the speculation information as the element information are executed.

Explanation of Signs

[0171] 3 User device (game system) 31 Control unit (computer) 35 Model processing unit (superiority / inferiority acquisition means, superiority / inferiority provision means, filter means, sorting means, information acquisition means, information provision means) 50 Battle screen (game screen) 55 Main deck placement area (deck area) 57 Hand placement area (hand area) CO Card object (object) CP Card placement area (card placement area) MO Display device CO5 Hand CO6 Deck PG2 Game program (computer program)

Claims

1. In a game including an opportunity to select a part of a plurality of options, a computer incorporated in a game system that provides a user with superiority / inferiority information regarding the superiority or inferiority of the influence exerted by each option before the selection of the part of the options, superiority / inferiority acquisition means for obtaining a plurality of pieces of superiority / inferiority information for each option of the at least one option by using a plurality of types of processes that respectively output the superiority / inferiority information for the at least one option, and superiority / inferiority providing means for providing the plurality of pieces of superiority / inferiority information for each option before the selection of the part of the options, A computer program configured to function as.

2. As the game, a battle-type game in which the user battles against an opponent is provided, The computer program according to claim 1, wherein the plurality of types of processes are configured to output, as the plurality of pieces of superiority / inferiority information, any information regarding the superiority or inferiority of the influence exerted by each option on victory in the battle-type game.

3. The computer program according to claim 2, wherein the plurality of types of processes are configured to output, as the plurality of pieces of superiority / inferiority information, any numerical information indicating the possibility of the progress becoming advantageous in the battle-type game through the magnitude of a numerical value.

4. The battle-type game is configured such that turns including the opportunity to select are alternately provided to the user and the opponent, The computer program according to claim 2, wherein the plurality of options include an option corresponding to a turn end instruction for ending the user's turn.

5. Each turn includes a plurality of phases that respectively provide a plurality of opportunities to select, The selection opportunities for each phase are configured to respectively provide, as the plurality of options, two or more options including different options among the selection opportunities for each phase, The computer program according to claim 4, wherein the two or more options for each selection opportunity each include an option corresponding to a phase end instruction for ending each phase.

6. The battle-type game is provided as a game in which the user battles against an opponent according to a predetermined rule via the plurality of objects of a display device that displays a plurality of objects. The computer program according to claim 4.

7. The competitive game is provided as a card game that uses a plurality of card objects as the plurality of objects, The card game is played using a plurality of card placement areas where the card objects to be used in the game against the opponent are to be placed respectively, a hand area where the hands as candidate card objects to be placed in each card placement area are to be placed, and a deck area where the deck as candidate card objects to be added to the hand is to be placed, The plurality of options include options corresponding to at least one of the plurality of card placement areas, the hands to be placed in each card placement area, and the addition of the deck to the hand. The computer program according to claim 6.

8. At least one of the plurality of types of processing is configured as a learned artificial intelligence model generated by causing a pre-learning model to perform machine learning on a predetermined learning dataset so as to output information on superiority or inferiority regarding a predetermined result in the game as the superiority or inferiority information. The computer program according to any one of claims 1 to 7.

9. All of the plurality of types of processing are configured as the learned artificial intelligence model, and are configured to output the plurality of superiority or inferiority information based on at least one difference among the predetermined result, the learning dataset, and the algorithm of the pre-learning model for learning the learning dataset. The computer program according to claim 8.

10. The superiority or inferiority providing means is provided with at least one of a filtering means for limiting the plurality of superiority or inferiority information to a part that satisfies a predetermined filtering condition, and a sorting means for sorting the plurality of superiority or inferiority information according to a predetermined sorting condition. The computer program according to any one of claims 1 to 7.

11. A game system in which a computer for providing a user with superiority or inferiority information regarding the superiority or inferiority of the influence given by each option in a game including a selection opportunity for selecting a part of a plurality of options is incorporated before the selection of the part of the options, for at least one of the plurality of options, The computer is a superiority or inferiority acquisition means for acquiring a plurality of superiority or inferiority information for each option of the at least one option by using a plurality of types of processing for outputting the superiority or inferiority information for each of the at least one option, and Before selecting a part of the options, a superiority / inferiority providing means for providing the plurality of superiority / inferiority information for each option. A game system that functions as.

12. In a game including a selection opportunity to select a part of a plurality of options, in a computer incorporated in a game system that provides a user with superiority / inferiority information regarding the superiority / inferiority of the influence given by each option before selecting the part of the options for at least one of the plurality of options, A procedure for obtaining a plurality of superiority / inferiority information for each option of the at least one option by using a plurality of types of processes that output the superiority / inferiority information for each of the at least one option, and A procedure for providing the plurality of superiority / inferiority information for each option before selecting the part of the options, A control method for causing the execution.

Images