Method for generating general-purpose pre-trained models

Abstract

To provide a system and a method for generating a universal learned model capable of appropriately controlling a group of operating devices having the same configuration.SOLUTION: A method for generating a universal learned model includes: an initial learning step S10 of subjecting a predetermined machine learning model to learning based on predetermined initial data to generate an initial learned model; an installation step S20 of incorporating the initial learned model that controls a predetermined operating device into a plurality of operating devices; an additional learning step S30 of individually performing additional learning based on respective operation data obtained by operating the respective operating devices; an integration processing step S40 of performing integration processing on an obtained plurality of individual learned models and transmitting them to an integration processing server; and an integration step S50 of obtaining the universal learned model.SELECTED DRAWING: Figure 4

Description

Technical Field

[0001] This invention relates to the control of an operating device using machine learning technology, particularly to the control of an operating device using a machine learning device having a tree structure.

Background Art

[0002] In recent years, due to the existence of a large amount of data, a dramatic improvement in the computing function of computers, the progress of algorithms, etc., the field of machine learning has attracted high attention. Among these, a new machine learning framework having a tree structure has been attracting attention (Patent Document 1).

[0003] FIG. 18 is an explanatory diagram showing the basic configuration of a learning tree according to a new machine learning framework. In FIG. 18(a), the structure of the learning tree in the learning method is shown, and in FIG. 18(b), an image of the state space corresponding to the structure is shown. As is clear from the figure, the learning tree structure is configured by arranging each node corresponding to each hierarchically divided state space in a tree shape or a lattice shape from the uppermost node (starting node or root node) to the lowermost node (terminal node or leaf node). Note that this figure shows an example in the case where N is 2, d is 2, and n is 2 in a learning tree of N layers, d dimensions, and n divisions, and the numbers 1 to 4 attached to the four terminal nodes in the first layer of the learning tree shown in FIG. 18(a) respectively correspond to the four state spaces shown in FIG. 18(b).

[0004] When performing learning processing using the above learning tree, the input data is sequentially associated with each divided state space and accumulates in those state spaces. At this time, when new data is input to a state space where there was no data until then, new nodes are sequentially generated. On the other hand, the predicted output is calculated by calculating a transition vector from each data included in the state space corresponding to the node corresponding to the new input data and taking the arithmetic mean of those transition vectors.

Prior Art Documents

Patent Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2016-173686 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Incidentally, in recent years there have been attempts to apply machine learning technology to various motion devices such as robotic arms in factories. For example, it is common practice to incorporate a pre-trained model into a group of motion devices that have the same configuration.

[0007] However, even with operating devices having the same configuration, variations in mechanical or electrical characteristics exist in individual devices due to individual differences and aging, which meant that the pre-trained model might not be able to properly control all of the operating devices.

[0008] The present invention was made against the technical background described above, and its objective is to provide a system, method, etc., for generating a general-purpose pre-trained model that can correspond to the individual characteristics of an operating device.

[0009] Further objects and effects of the present invention will be readily apparent to those skilled in the art by referring to the following description in the specification. [Means for solving the problem]

[0010] The technical challenges described above can be solved by a method or system having the following configuration.

[0011] The method for generating a general-purpose pre-trained model according to this disclosure comprises: an initial training step of generating an initial pre-trained model by training a predetermined machine learning model based on predetermined initial data; and an integration step of integrating a plurality of individual pre-trained models obtained by incorporating the initial pre-trained model, which controls a predetermined operating device, into a plurality of the operating devices, and performing additional training on each of the individual operating devices based on individual operation data obtained by operating each of the operating devices, in order to obtain a general-purpose pre-trained model.

[0012] With this configuration, after performing common initial training, additional training is performed to suit each individual operating device, and these are then integrated, thereby generating a general-purpose pre-trained model that can handle the specific characteristics of a given operating device. Furthermore, because initial training is performed, a certain level of accuracy can be guaranteed, allowing for the generation of a general-purpose pre-trained model while continuing to use the operating device.

[0013] The generation of the initial trained model may be performed on a predetermined initial training server connected to a network, and the incorporation of the initial trained model into the operating device may be performed by downloading it from the initial training server to each operating device via the network.

[0014] Furthermore, the initial training model may be incorporated into the operating device by mounting the initial training model onto an integrated circuit and then incorporating the integrated circuit into each of the operating devices.

[0015] The method for generating the general-purpose trained model may further include a re-integration step of incorporating the general-purpose trained model into each of the aforementioned operating devices.

[0016] The aforementioned machine learning model may be a learning model having a tree structure constructed by hierarchically arranging multiple nodes, each corresponding to a hierarchically divided state space.

[0017] The integrated circuit may include a communication unit that communicates with a predetermined integrated processing server, the individually trained models may be transmitted from each operating device to the integrated processing server via each communication unit, and the integrated processing may be performed on the integrated processing server.

[0018] The initial data may be data generated based on the test operating device or a simulation of the operating device.

[0019] The integration step may further include a selective integration step that selectively integrates the individually trained models based on the accompanying information corresponding to each of the individually trained models.

[0020] Furthermore, this disclosure can also be conceived as the following method. That is, the method relating to this disclosure comprises: an initial learning step of generating an initial trained model by training a predetermined machine learning model based on predetermined initial data; an integration step of incorporating the initial trained model, which controls a predetermined operating device, into a plurality of the operating devices; an additional learning step of generating a plurality of individual trained models by performing additional learning on each of the operating devices based on individual operation data obtained by operating each of the operating devices; and an integration step of obtaining a general-purpose trained model by integrating each of the individual trained models.

[0021] Furthermore, this disclosure can also be conceived as a system. That is, the system according to this disclosure comprises: an initial learning unit that generates an initial trained model by performing training on a predetermined machine learning model based on predetermined initial data; and an integration unit that incorporates the initial trained model, which controls predetermined operating devices, into a plurality of the operating devices, and performs additional training on each of the individual trained models obtained by operating each of the operating devices, and integrates these individual trained models to obtain a general-purpose trained model. [Effects of the Invention]

[0022] According to the present disclosure, a general trained model that can also accommodate individual characteristics of operating devices can be generated.

Brief Description of Drawings

[0023] [Figure 1] FIG. 1 is an explanatory diagram regarding the overall configuration of the system. [Figure 2] FIG. 2 is an explanatory diagram regarding the configuration of the information processing device. [Figure 3] FIG. 3 is an explanatory diagram regarding the configuration of the operating device. [Figure 4] FIG. 4 is a general flowchart regarding the generation of a general trained model. [Figure 5] FIG. 5 is a general flowchart regarding the learning process. [Figure 6] FIG. 6 is a flowchart regarding the calculation process of errors. [Figure 7] FIG. 7 is a flowchart regarding the operation of the operating device. [Figure 8] FIG. 8 is a flowchart regarding the prediction process. [Figure 9] FIG. 9 is a flowchart regarding the integration process. [Figure 10] FIG. 10 is a detailed flowchart regarding the integration process. [Figure 11] FIG. 11 is a conceptual diagram (Part 1) regarding the integration process. [Figure 12] FIG. 12 is a conceptual diagram (Part 2) regarding the integration process. [Figure 13] FIG. 13 is an overall configuration diagram of the system (Second Embodiment). [Figure 14] FIG. 14 is a general flowchart regarding the generation of a general trained model (Second Embodiment). [Figure 15] FIG. 15 is a flowchart regarding the selection process of the trained models to be integrated. [Figure 16] FIG. 16 is a conceptual diagram of the integration process based on the accompanying information. [Figure 17]Figure 17 is a conceptual diagram of a multi-stage integration process performed using supplementary information. [Figure 18] Figure 18 is an explanatory diagram of the basic structure of a learning tree. [Modes for carrying out the invention]

[0024] Hereinafter, one embodiment of the present invention will be described in detail with reference to the attached drawings.

[0025] <1. First Embodiment> A first embodiment of this disclosure will be described with reference to Figures 1 to 13.

[0026] <1.1 Device configuration> Referring to Figures 1 to 3, the device configuration of the general-purpose trained model generation system 5 according to this embodiment will be described.

[0027] Figure 1 is an explanatory diagram of the overall configuration of the general-purpose pre-trained model generation system 5 according to this embodiment. As is clear from the figure, the general-purpose pre-trained model generation system 5 has an information processing device 1 that performs information processing such as the learning process described later, an integration processing server 2 that integrates pre-trained models, and a plurality of operating devices 3 such as a robot arm, and each of the devices 1 to 3 is connected to each other via the Internet. As will be described later, a predetermined pre-trained model that has been processed by the information processing device 1 is mounted on the integrated circuit 4, and the integrated circuit after the pre-trained model is mounted is mounted on each of the operating devices 3.

[0028] Figure 2 is an explanatory diagram of the schematic configuration of the information processing device 1. As is clear from the figure, the information processing device 1 has a display unit 11 including a display, an audio output unit 12 that outputs sound via a speaker, an input unit 13 that accepts input from a keyboard, mouse, etc., a control unit 14 consisting of a CPU, a storage unit 15 consisting of memory such as ROM, RAM, flash memory, and hard disk for storing operating programs and predetermined data, and a communication unit 1 for communication with external devices, and these are connected to each other via a system bus. The control unit 14 reads programs and predetermined data from the storage unit 15 and performs predetermined information processing, including machine learning. Note that the hardware configuration of the integrated processing server 2 is substantially the same as that of the information processing device 1, so the description of its configuration is omitted here.

[0029] Figure 3 is an explanatory diagram of the configuration of the integrated circuit 4 mounted inside the operating device 3. As is clear from the figure, the integrated circuit 4 consists of a control unit 41 consisting of an MPU or the like, a storage unit 42 consisting of a ROM, RAM or the like, a communication unit 43 consisting of a communication unit for communicating with external devices such as the integrated processing server 2, an output unit 44 that is connected to the operating circuit of the operating device 3 and provides control output, and an I / O unit 45 that performs input and output with external devices, and these are electrically connected to each other via a system bus or the like. In other words, the control unit 41 can control a predetermined operating device 3 using a learned model read from the storage unit 42.

[0030] <1.2 Operation> Next, the operation of the general-purpose pre-trained model generation system 5 according to this embodiment will be described with reference to Figures 4 to 12.

[0031] <1.2.1 Initial Learning Steps> Figure 4 is a general flowchart of the method for generating a general-purpose pre-trained model. As is clear from the figure, once processing begins, the information processing device 1 performs the training process for a model having a tree structure (S10).

[0032] Figure 5 is a detailed flowchart of the learning process (S10). As is clear from the figure, when the process starts, the initial data is first read and input to the learning model is performed (S101). Here, the initial data consists of input data and corresponding training data, and the training data is a set of actual data obtained by actually operating the operating device 3 or a set of data obtained by simulation using a physical simulator, etc. For example, if the operating device 3 is a robot arm, depending on the control mode, the command value of the output torque, actual joint angle data obtained by the angle sensor, the end-effector position captured by the camera, etc., can be used as training data. Once the input process to the learning model is completed, the state space to which the input data should belong and the node corresponding to that state space are identified, and a path from the root node to the terminal node is identified (S102). At this time, if there is a node on the path that has never been active before, a new node is generated at that position. After that, for each node on the path, an error calculation process corresponding to the prediction error of each node is performed (S103). Subsequently, the process (S104) is performed to save all data and parameters for each node as a file, and the process ends. Here, parameters refer to information such as the range of values ​​to be learned (learning space), i.e., the maximum and minimum values ​​of the input and output data, and the network structure (number of layers, number of dimensions, number of divisions).

[0033] Figure 6 is a flowchart showing the details of the error calculation process (S103). When the process starts, the root node is first set as the node of interest (S1031). Next, for the node of interest, the error is updated using the formula Et+1=(1-α)Et+α|O-O'| (S1032). Here, Et+1 is the error after the update, Et is the error before the update, and α is the forgetting coefficient (0<α<1). As is clear from the formula, the error after the update Et+1 is calculated by weighting the absolute difference between the error before the update Et and the predicted output O' based on the data included in the node of interest, using the forgetting coefficient α. In other words, the smaller the difference between the predicted output O based on the input data and the predicted output O' based on past data included in the node of interest, the smaller Et+1 will be. Therefore, the smaller the error Et+1, the better the prediction accuracy. The updated error Et+1 is associated with the node of interest and stored in the memory unit 15 as error E (S1033).

[0034] Subsequently, a process is performed to determine whether the node of interest is the lowest-level node (S1034). If it is not the lowest-level node, the node of interest is changed to the next lower node (S1035). This series of processes—changing the node of interest to the next lower node on the path (S1035), updating the error (S1032), and saving the error (S1033)—is repeated until the lowest-level node is reached (S1034NO). On the other hand, if the node of interest reaches the lowest-level node (end node) (S1034YES), the process ends. In other words, this series of processes generates an error E equivalent to the prediction accuracy for each node on the path.

[0035] The method for calculating the error E is not limited to the above. Therefore, for example, errors obtained by referring to nodes on the path below the node of interest may also be used.

[0036] <1.2.2 Installation Steps> Returning to Figure 4, once the learning process is complete, the trained model corresponding to the learning result is loaded onto the integrated circuit (IC) 4, and then the integrated circuit 4 is loaded onto each operating device 3 (S20). Specifically, the tree structure model trained based on the initial data described above is read by the control unit 41 and stored in the storage unit 42 in an executable manner. Furthermore, the trained model is configured to allow for additional learning, and is configured so that when new data is obtained in each operating device 3, further additional learning can be performed based on that data.

[0037] <1.2.3 Additional Learning Steps> After the installation process on the operating device 3 is completed, the operating device 3 is actually controlled using the installed trained model, and additional training is performed based on the actual operation data obtained from this operation (S30).

[0038] Figure 7 is a detailed flowchart of the operation of the actuator 3 (S30). As is clear from the figure, once processing begins, the operation of the actuator 3 is controlled by a predictive process using a trained model (S301).

[0039] Figure 8 is a flowchart detailing the prediction process (S301). As is clear from the figure, when the process starts, new data related to the operating device 3 is first acquired and used as input data (S3011). Next, based on the input data, the state space (node) to which it should belong is identified, thereby identifying a path consisting of a series of nodes (S3012). Once a path is identified, the node with the minimum error E on that path is identified (S3013), and a prediction output is generated based on the arithmetic mean (composite vector of transition vectors) of the transition vectors of each data corresponding to the node with the minimum error E (S3014). This prediction output is transmitted as a command value to the operating circuit of the operating device 3 via the output unit 44, and the operating device 3 operates accordingly. In other words, a prediction output is generated based on the node with the minimum error on the path, thereby enabling the generation of a prediction output based on nodes where the prediction accuracy is expected to be relatively good.

[0040] In this embodiment, the prediction output was generated based on data corresponding to the node with the smallest error along the path, but the process is not limited to this. Therefore, for example, the configuration may always generate the prediction output based on data corresponding to the terminal node.

[0041] Returning to Figure 7, once the prediction process is complete, the actual operation data of the operating device 3 is stored (S302). This series of processes, including the prediction process (S301) and the operation data storage process (S302), is repeated until a predetermined operation termination condition is met (S303NO). When the predetermined operation termination condition is met (S303), the series of operations ends, and an additional learning process is performed using the stored operation data as training data (S304). This additional learning process is substantially the same as the process shown in Figure 2 (S101~S104), so its explanation is omitted here. After that, the process ends.

[0042] To explain the operation in Figure 7 in more detail, we will describe an example task in which a single-joint robot arm, whose initial joint angle is 0 degrees relative to a predetermined standard, is changed to a state of 30 degrees. First, when processing starts, the initial angle (0 degrees) and the target angle (30 degrees) are given to a trained model that has learned the relationship between them and the command value of the output torque. Then, a prediction process for the command value of the output torque is performed, and the robot arm is controlled according to the predicted output (S301). After the motion control, the sensor provided on the robot arm detects the angle data as actual motion data (S302). After that, when the task is completed by the actual driving of the robot arm, the operation ends (S303YES), and additional learning processing is performed using the actual angle value obtained from the sensor (S304). As described above, the robot arm performs additional learning for each actual movement to improve accuracy. Here, a configuration is used in which the output torque is predicted using the initial angle and target angle as input, but this disclosure is not limited to such a learning configuration. Therefore, it goes without saying that other combinations of input and output configurations may be adopted.

[0043] <1.2.4 Integration Processing Steps> Returning to Figure 4, once the individual additional training process (S30) in each operating device 3 is completed, each operating device 3 sends data related to the additionally trained model, i.e., data related to nodes and parameters, to the integrated processing server 2 via the network (S40). Upon receiving this data, the integrated processing server 2 performs a process to integrate the additionally trained models (S50).

[0044] Figure 9 is a detailed flowchart of the integration process (S50). When the process starts, the parameter files generated when creating the trained models are read for two different trained models (S501). Based on these parameter files, it is determined whether the training intervals match or approximately match (S502). If they do not match (S502NO), an error message is output to the user via a display (not shown) or the like informing them that integration cannot be performed (S505), and then it is determined whether the integration process of all training trees has been completed (S506). On the other hand, if the training spaces match or approximately match, the data files for the nodes of the two trained models are read (S503). After that, the integration process of the training trees related to the two different trained models, which will be described later, is performed (S504), and then it is determined whether the integration process of all training trees has been completed (S506). If it is determined that the integration process of all training trees has not been completed (S506NO), the integration process is performed with a different combination of training trees (S506NO). On the other hand, if it is determined that the integration process of all learning trees has been completed (S506YES), the process terminates. Here, approximate agreement of the learning spaces means, for example, that the maximum and minimum values ​​of the data in one trained model are included within a range that allows for a predetermined margin (e.g., about 10%) between the maximum and minimum values ​​of the data in one trained model.

[0045] Figure 10 is a detailed flowchart of the integration process (S504). When the process starts, the first node position is set as the position of interest (S5041). Next, a determination process is performed to determine whether a node exists in both learning trees at the position of interest (S5042). If a node exists in both learning trees at the position of interest (S5042YES), the integration process of those two nodes is performed (S5043).

[0046] Here, we will describe the node integration process according to this embodiment. Figure 12 is a conceptual diagram of the node integration process (S5043). Figure 12(a) is a first example of integration when two colored nodes on the upper left and right are integrated into one colored node at the bottom. In this figure, E represents the error mentioned above, and C represents the number of data points included so far. The error E of the upper left node is α (>0), and the number of data points C is k (an integer greater than or equal to 0), while the error E of the upper right node is β (>0), and the number of data points C is m (an integer greater than or equal to 0). When integrating these nodes, the data associated with each node is integrated so that it is associated with one node. Furthermore, the error E after integration is calculated by weighting and adding the above-mentioned errors α and β using the respective number of data points k and m. Furthermore, the number of data points C after integration is calculated by adding the respective number of data points k and m together.

[0047] With this configuration, node integration is performed by weighting and summing according to the number of data points included, thus enabling the integration of learning-related parameters based on reliability derived from the number of data points.

[0048] Figure 12(b) shows a second example of merging two colored nodes at the top left and right into one colored node at the bottom. The error E of the top left node is α (>0), and the number of data points C is n (a non-negative integer), while the error E of the top right node is ∞, and the number of data points C is 1. Here, the error E is ∞ because the number of data points associated with that node is still 1, and the error is not yet defined, so this value is assigned for convenience. When merging these nodes, the data associated with each node is merged so that it is associated with one node. The error E after merging is calculated as α, based on the above-mentioned errors α and ∞. Furthermore, the number of data points C after merging is calculated as n+1 by adding the number of data points n and 1.

[0049] In this embodiment, when integrating nodes, the data associated with each node, the error E, and the number of data points C are included in the integration process. However, the objects of integration are not limited to these parameters. Therefore, other parameters associated with nodes may also be included in the integration process.

[0050] Returning to Figure 10, once the node integration process (S5043) is complete, the integrated nodes are stored in the memory unit (S5044). Subsequently, a determination process is performed to determine whether or not all node locations have been processed (S5045).

[0051] On the other hand, if there is no node at the position of interest in either learning tree (S5042NO), then a determination is made to determine whether or not a node exists in either learning tree (S5047). If there is no node in either learning tree (S5047NO), then no further processing is performed, and a determination is made to determine whether or not all node positions have been processed (S5047NO). If a node exists at the position of interest in either learning tree (S5047YES), the nodes below the position of interest in the data of the existing learning tree are copied and saved as the merged data (S5048). After that, a determination is made to determine whether or not all node positions have been processed (S5045).

[0052] With this configuration, if a node exists only in one trained model, then the nodes below that node also exist only in that trained model, enabling efficient integration processing.

[0053] If processing has been performed for all node positions (S5045YES), the process ends. On the other hand, if processing has not yet been performed for all node positions (S5045NO), the focus position is changed to the next predetermined node position (S5046), and then node merging (S5043) or duplicating one of the nodes (S5048) is performed again. At this time, the focus position is set in order from the highest node position (depth-first search). This is because if a node exists in only one of the learning trees (S5047YES), it is necessary to duplicate the nodes below the focus node (S5048), and searching from the highest node position is more efficient.

[0054] Figure 11 is a conceptual diagram of the merging of learning trees. As is clear from the figure, when the first learning tree in the upper left of the figure and the second learning tree in the upper right of the figure are merged, the black nodes undergo the merging process, as shown in the lower part of the figure, while the white nodes are duplicates of the nodes in either the first or second learning tree.

[0055] With this configuration, since it involves integrating learning models based on tree structures, the integration process becomes easier compared to neural networks and the like.

[0056] Furthermore, with this configuration, the individual trained models from the numerous operating devices 3 that have undergone additional training can be aggregated and integrated into the integrated processing server 2, making it possible to generate a trained model that can be used universally for that type of operating device.

[0057] <1.2.5 Download Steps> Returning to Figure 4, once the integration processing of the trained model in the integration processing server 2 (S50) is completed, the integrated model (general-purpose trained model) is downloaded from the integration processing server 2 to each operating device 3 via the network (S60). Specifically, after a predetermined period of time has elapsed, the owner of the operating device 3 downloads the integrated model to an information processing device such as a personal computer (not shown), and then transfers the integrated model from the information processing device to each operating device 3 via the I / O unit 45 to be stored in the respective storage units 42 of the operating device 3. After that, the process ends.

[0058] With this configuration, the actuator 3 can be controlled with greater precision using an integrated model that has learned the characteristics of various actuators 3. Furthermore, the owner of the actuator 3 can choose to use either the pre-trained model already installed or the integrated model.

[0059] In this embodiment, the owner of the operating device 3 downloads the integrated model to an information processing device such as a personal computer (not shown). However, this disclosure is not limited to such a configuration. Therefore, for example, the owner of the operating device 3 may issue a download command to the operating device 3, and the operating device 3 may download the latest integrated model via the communication unit 43.

[0060] <2. Second Embodiment> Next, a second embodiment of the present invention will be described with reference to Figures 13 to 16. In this embodiment, various supplementary information is used when performing integration processing on trained models.

[0061] Figure 13 is an overall configuration diagram of the general-purpose trained model generation system 100 according to this embodiment. As is clear from the figure, the general-purpose trained model generation system 100 is configured by connecting a first information processing device 101, a second information processing device 102, and an information processing terminal 103 via a network such as the Internet. Each device exchanges information with the others via communication.

[0062] The first information processing device 101, as described below, is mainly responsible for learning processing, performing initial learning and integration processing of trained models. The second information processing device 102 functions as an application server, handling tasks such as providing a predetermined application program to the information processing terminal 103 and performing predetermined server-side processing, and also communicating information with the first information processing device 101 and the information processing terminal 3. The information processing terminal 103 is an information processing device such as a smartphone or tablet computer, and can download a predetermined application program from the second information processing terminal 102. The information processing terminal 103 works in cooperation with the second information processing device 102 to provide the predetermined application to the user. The information processing terminal 103 is connected to a wearable terminal 104 that can be paired using standards such as Bluetooth (registered trademark). The wearable terminal 104 is equipped with various sensors that sense blood pressure, heart rate, body temperature, steps, etc., and the acquired sensor information is transmitted to the information processing terminal 3. Note that the configuration of each device is substantially the same as the configuration shown in Figure 2, so it is omitted here.

[0063] Here, although various examples of application program content are conceivable, in this embodiment, the application program takes various sensor information acquired from the wearable terminal 104 as input and predicts a predetermined disease risk for the user of the wearable terminal 104. The information acquired from the sensors includes vital sign information such as blood pressure, heart rate, and body temperature, as well as activity level information such as the number of steps taken.

[0064] Figure 14 is a general flowchart for generating a general-purpose pre-trained model according to this embodiment.

[0065] When processing begins, the first information processing device 101 performs initial learning using a learning model having the same tree structure as in Embodiment 1 (S81). The details of the learning process are substantially the same as those in Figures 5 and 6, so a detailed explanation is omitted.

[0066] However, the learning process in this embodiment is performed by supervised learning, using vital information such as blood pressure, heart rate, and body temperature, and activity level information such as step count as input, and using actual disease information of various users (various numerical values, diagnostic results, health status evaluations, etc.) as training data. In other words, the learning process makes it possible to obtain a trained model that outputs disease information such as the predicted disease risk for the user based on vital information and activity level information.

[0067] The trained model generated in the first information processing device 101 is downloaded to each information processing terminal 103 via application software running on the information processing terminal 103 (S82). Alternatively, it may be downloaded to each information processing terminal 103 in response to a command from information processing terminal 102, which functions as an application server.

[0068] At this time, the information processing terminal 103 accepts input of supplementary information about the user via an application program and stores it in its internal memory. Supplementary information includes, for example, information about age, gender, or the area where the user lives.

[0069] Subsequently, additional learning is performed individually on each information processing terminal 103 (S83). The details of the additional learning process are substantially the same as those shown in Figures 7 and 8, so a detailed explanation is omitted. However, in the additional learning process in this embodiment, the actual disease information of each user, input via the application software, is used as the actual operation data. Furthermore, the end of the operation (S303) is determined, for example, by the elapsed time of a predetermined period.

[0070] After a predetermined period has elapsed, the newly trained model and accompanying information concerning the user who will transmit the trained model are transmitted to the first information processing device 101 (S84).

[0071] Upon receiving the additionally trained model and accompanying information, the first information processing device 101 performs a process to integrate the trained models based on the accompanying information (S85).

[0072] The details of the integration process are substantially the same as those shown in Figures 9 to 12, so a detailed explanation will be omitted. However, in the integration process in this embodiment, a process for selecting the trained model to be integrated is performed prior to the integration process shown in Figure 9.

[0073] Figure 15 is a flowchart relating to the process of selecting trained models to be integrated. As is clear from the figure, when the process starts, the first information processing device 101 selects trained models to be integrated according to predetermined selection criteria and received supplementary information (S851). Here, the predetermined selection criteria are selection criteria that are stored in the first information processing device 101 in advance or set by other devices or terminals, and their contents may be, for example, information about the area in which the user lives. That is, the first information processing device 101 performs a process to select only trained models relating to users who live in the same area, for example.

[0074] Subsequently, the selected pre-trained models are integrated as appropriate to obtain multiple general-purpose pre-trained models (S852). That is, each general-purpose pre-trained model is an integration of pre-trained models relating to users residing in the same region. The integration process is substantially the same as that shown in Figure 9, so its explanation is omitted.

[0075] Figure 16 is a conceptual diagram of the integration process (S85) based on supplementary information. In this figure, multiple trained models with a tree structure are arranged in two layers, upper and lower. The lower layer (first layer) consists of multiple trained models 110-1 to 110-N (where N is a natural number greater than or equal to 2) that have undergone additional training individually at each information processing terminal 103. The upper layer (second layer) consists of multiple trained models 120-1 to 120-M (where M is a natural number greater than or equal to 2) with a tree structure, generated by integrating the trained models 110-1 to 110-N in the lower layer. As is clear from this figure, for example, if a predetermined selection criterion is the region in which a user lives, then multiple trained models with a tree structure, as shown in the upper layer (second layer), will be generated by integrating multiple trained models based on the multiple trained models in the lower layer obtained from users living in the same region.

[0076] Returning to Figure 14, once the integration process (S85) is complete, a process is performed to store each integrated trained model, and the process ends (S86). The stored trained models may then be configured to be downloadable to each information processing terminal 103. Alternatively, the stored trained models may be transmitted to the second information processing device. At this time, the system may be configured to allow downloading of a corresponding trained model based on associated information. For example, a system may be configured to download a predictive model (trained model) suitable for individuals living in the same region as the user.

[0077] With this configuration, additional trained models are integrated based on predetermined selection criteria and associated information, making it possible to generate a general-purpose trained model by integrating trained models with common associated information. This allows for the generation of trained models suitable for specific applications or users.

[0078] <3. Variant> In the first embodiment, it was described that the trained model after additional training is transmitted from the operating device 3, but this disclosure is not limited to such a configuration. Therefore, for example, integration may be achieved by transmitting operation data from the operating device 3 to the integration processing server 2 and training the operation data in the integration processing server 2.

[0079] In the embodiments described above, there are no particular limitations on the timing of sending the trained model after additional training. However, the timing of the sending may be after a certain amount of additional training has progressed, or it may be sequential.

[0080] In the above embodiment, information is aggregated to an integrated processing server 2 on the network, but it may also be aggregated to, for example, a PLC (Programmable Logic Controller) connected to the network.

[0081] In the first embodiment, a robotic arm was exemplified as the operating device 3, but the invention is not limited to such an operating device. Therefore, it can be applied to a variety of operating devices, such as sensor equipment, mobile phones, automobiles, drones, AI speakers, ships, offshore floats, tunnel boring shields, construction heavy machinery, railways, EVs, air conditioners, compressors, factory machinery, household robots, nuclear fusion control, PLCs, communication gateways, smart homes, wearable terminals, home appliances, various meters (electricity, water, gas, etc.), game consoles, etc. It should be noted that the operation described in this disclosure is not limited to the mechanical operation of the device, but also includes, for example, electrical or electromagnetic operation in the device.

[0082] In the first embodiment, a pre-trained model is mounted on an integrated circuit 4, and the integrated circuit 4 is incorporated into an operating device 3, after which additional training is performed individually. However, this disclosure is not limited to such a configuration. Therefore, for example, after generating a pre-trained model by performing pre-training on the information processing device 1, the pre-trained model may be downloaded directly or indirectly to each operating device 3, and additional training may be performed in each operating device 3 based on the downloaded pre-trained model.

[0083] In the first embodiment, the information processing device 1 and the integrated processing server 2 are shown as separate configurations, but they may be integrated. Furthermore, some functions may be independent, or distributed processing may be performed.

[0084] The integrated circuit 4 in the first embodiment may be configured as an FPGA (Field-Programmable Gate Array) or the like.

[0085] In the second embodiment, the integration process based on supplementary information was described as being performed in two stages, as shown in Figure 16. However, the present invention is not limited to such a configuration. Therefore, for example, the system may be configured to perform integration in multiple stages, store each stage, and provide an appropriate integrated and trained model in response to a request from an information processing terminal 103 or the like.

[0086] Figure 17 is a conceptual diagram of a multi-stage integration process performed using supplementary information. The bottom row shows the first layer 130, consisting of trained models 130-1 to 130-N obtained by additional learning on individual information processing terminals 103. The middle row shows the second layer 140, consisting of trained models 140-1 to 140-M obtained by integrating the trained models of the layer immediately below. The top row shows the Kth layer 150, consisting of trained models 150-1 to 150-L obtained by integrating the trained models of the layer immediately below. As shown in the figure, multi-stage integration processing may be performed based on supplementary information. For example, in the second layer, the trained models of the first layer may be integrated based on regional information, and in the third layer, integration processing may be performed based on regional information from a broader perspective. In addition, in other layers, integration processing may be performed stepwise based on gender, age information, etc.

[0087] With this configuration, it is possible to provide the information processing terminal 103 and the second information processing device 102 with an appropriate integrated and trained model as needed.

[0088] In the embodiments described above, a learning model using a tree structure was adopted as the learning model, but the method is not limited to this. Therefore, other integrable learning models can be adopted. [Industrial applicability]

[0089] This system can be used in industries that manufacture or otherwise produce general-purpose pre-trained models. [Explanation of symbols]

[0090] 1. Information Processing Device 2. Integrated Processing Server 3 Operating device 4. Integrated Circuits 5. General-purpose pre-trained model generation system 100 General-Purpose Pre-Trained Model Generation Systems 101 First Information Processing Device 102 Second Information Processing Device 103 Information Processing Terminal 104 Wearable devices

Claims

1. A trained model generation step involves performing machine learning processing on each of the multiple operating devices based on the data obtained from each of them to generate a trained model corresponding to each of the said operating devices. An integration step of selectively integrating the trained models based on the accompanying information corresponding to each of the trained models to generate an integrated trained model to be provided to each of the operating devices, Equipped with, A method wherein each of the aforementioned incidental information is attribute information of the user of the corresponding operating device.

2. The integration step involves incorporating the integrated learned model into each of the aforementioned operating devices. The method according to claim 1, further comprising:

3. The method according to claim 1, wherein each of the operating devices is configured to allow selection whether or not to apply the integrated trained model to each of the operating devices.

4. The method according to claim 1, wherein the integration process is a multi-stage integration process that also includes the integration of integrated trained models.

5. The integration process is a process of integrating each of the trained models that have common corresponding ancillary information, The method according to claim 1, wherein the integrated pre-trained model is a pre-trained model that can be used in general by the user having the common attribute information.

6. The attribute information includes a plurality of different attribute information, The method according to claim 4, wherein, in the multi-stage integration process, attribute information that forms the basis for integration decisions is set for each hierarchical level.

7. A trained model generation unit performs machine learning processing on each of the multiple operating devices based on the data obtained from each of them to generate a trained model corresponding to each of the operating devices. An integration unit selectively integrates the learned models based on the accompanying information corresponding to each of the learned models to generate an integrated learned model to be provided to each of the operating devices. Equipped with, Each of the aforementioned incidental information is attribute information of the user of the corresponding operating device.

8. A trained model generation unit performs machine learning processing on each of the multiple operating devices based on the data obtained from each of them to generate a trained model corresponding to each of the operating devices. An integration unit selectively integrates the learned models based on the accompanying information corresponding to each of the learned models to generate an integrated learned model to be provided to each of the operating devices. Equipped with, Each of the aforementioned incidental information is attribute information of the user of the corresponding operating device, in a system.

Images