Prediction program, prediction method and information processing unit

Abstract

To reduce a space calculation amount.SOLUTION: An information processing unit inputs first data to a first machine learning model to calculate a first feature quantity. The information processing unit specifies information for updating parameters of the first machine learning model based upon the first feature quantity. The information processing unit uses the updating information to update the parameters of the first machine learning model. The information processing unit outputs the prediction result obtained by inputting the first data to the first machine learning model having the parameters updated.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a prediction program, a prediction method, and an information processing device. [Background technology]

[0002] Specialist LLMs (Large Language Models) specialized in a particular field are generated by additional training on general-purpose LLMs (LLMs) trained using general training data. For example, a general-purpose LLM is additionally trained using training data related to mathematics to generate a mathematics-specific LLM, and a general-purpose LLM is additionally trained using training data related to program coding to generate a coding-specific LLM.

[0003] An information processing device is known that uses the multiple expert LLMs generated in this way to make predictions on data to be predicted. For example, the information processing device inputs the data to be predicted into a classifier, assigns the data to an appropriate expert LLM based on the results of the classifier, and makes predictions using the expert LLM. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2023-73095 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the above technique requires the use of a classifier, which increases the amount of space required for computation, such as the training data required to generate the classifier, the memory capacity required to store the classifier, and the need to retrain the classifier every time the number of expert LLMs is increased.

[0006] In one aspect, an object of the present invention is to provide a prediction program, a prediction method, and an information processing device that can reduce the amount of spatial calculation. [Means for solving the problem]

[0007] In a first proposal, the prediction program causes a computer to execute the following process: input first data into a first machine learning model to calculate a first feature; identify information for updating parameters of the first machine learning model based on the first feature; update the parameters of the first machine learning model using the updating information; and input the first data into the first machine learning model with the updated parameters, thereby outputting the prediction result obtained. [Effects of the Invention]

[0008] According to one embodiment, the space complexity can be reduced. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an information processing apparatus according to a first embodiment. [Figure 2] FIG. 2 is a diagram illustrating the reference technology. [Figure 3] FIG. 3 is a functional block diagram of the information processing apparatus according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating the difference information DB. [Figure 5] FIG. 5 is a diagram for explaining the generation of the difference information DB. [Figure 6] FIG. 6 is a diagram illustrating the calculation process at the time of prediction. [Figure 7] FIG. 7 is a diagram illustrating the update process during prediction. [Figure 8] FIG. 8 is a diagram illustrating the prediction process at the time of prediction. [Figure 9] FIG. 9 is a flowchart showing the flow of the pre-learning process. [Figure 10] FIG. 10 is a flowchart showing the flow of the prediction process. [Figure 11]FIG. 11 is a diagram illustrating an example of processing using a plurality of weight differences. [Figure 12] FIG. 12 is a diagram illustrating another example of prediction processing using an updated general-purpose LLM. [Figure 13] FIG. 13 is a diagram illustrating an example of updating the general-purpose LLM. [Figure 14] FIG. 14 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0010] The following describes in detail embodiments of the prediction program, prediction method, and information processing device disclosed herein with reference to the accompanying drawings. Note that the present invention is not limited to these embodiments. The embodiments can be combined as appropriate within a consistent range. [Example]

[0011] (Description of information processing device) Fig. 1 is a diagram illustrating an information processing device 10 according to a first embodiment. The information processing device 10 illustrated in Fig. 1 is an example of a computer that selects an appropriate large language model (LLM) for an input prompt, which is an example of data to be predicted, and predicts the next token using the selected large language model.

[0012] In general, when there are multiple trained expert LLMs (LLMs) trained specifically for a particular domain, it is possible to improve overall accuracy by using the most appropriate expert LLM for the input prompt. For example, if a math prompt is input, accuracy can be improved by routing it to an LLM trained specifically for math.

[0013] Here, a reference technique for routing will be described. Figure 2 is a diagram illustrating the reference technique. As shown in Figure 2, in the reference technique, the memory of a GPU (Graphics Processing Unit) stores a general-task LLM trained using general learning data, a mathematics-specific LLM generated by performing additional learning on the general-task LLM using learning data related to mathematics, and a coding-specific LLM generated by performing additional learning on the general-task LLM using learning data related to coding. The GPU memory also stores a classifier that predicts the LLM to be used for an input prompt, which has been trained as a router to predict which model is likely to have a high accuracy rate for a prompt.

[0014] Consider the prediction process for the prompt "When x+y=22 and x=12, what is y?" in this situation. In this case, the prompt "When x+y=22 and x=12, what is y?" is input to the classifier, and after the classifier determines the destination to be the mathematics-specialized LLM, the classifier inputs the prompt to the mathematics-specialized LLM. The mathematics-specialized LLM then outputs a prediction result based on the input prompt.

[0015] As described above, the reference technology uses a dedicated routing model (classifier) ​​for routing, but the training cost of the routing model and the space computational complexity required to deploy the routing destination model are high. For example, a huge amount of training data is required to generate the classifier, and the classifier needs to be retrained every time the number of routing destination expert LLMs increases, requiring GPU memory to store the classifier and multiple expert LLMs.

[0016] Therefore, the information processing device 10 according to the first embodiment uses the embedding feature of the general-purpose LLM itself to calculate what kind of data the input prompt is similar to in terms of the embedding feature, and predicts the next token by changing the weight of the general-purpose LLM so that it approaches the weight of the expert LLM (specialized LLM) associated with similar data.

[0017] Specifically, as shown in FIG. 1, the information processing device 10 maintains a difference information DB that associates the features of prompts for each expert LLM with the differences in network weights (parameters) between each expert LLM and a general-purpose LLM. In this state, the information processing device 10 inputs a prompt, which is an example of first data to be predicted, into a general-purpose LLM, which is an example of a first machine learning model, to calculate embedded features. Next, the information processing device 10 searches the difference information DB for features similar to the calculated embedded features and identifies the difference information of the general-purpose LLM. Next, the information processing device 10 updates the parameters of the general-purpose LLM using the identified difference information. Thereafter, the information processing device 10 inputs the prompt into the general-purpose LLM with the updated parameters, and predicts and outputs the next token.

[0018] In this way, the information processing device 10 according to the first embodiment uses the generalized LLM itself for domain identification, and reduces the amount of spatial calculation by changing the weights (parameters) of the network of the generalized LLM itself.

[0019] (Functional configuration of information processing device) 3 is a functional block diagram illustrating a functional configuration of the information processing device 10 according to Example 1. As illustrated in FIG.

[0020] The communication unit 11 is a processing unit that controls communication with other devices, and is realized by, for example, a communication interface, etc. For example, the communication unit 11 receives a prompt, which is an example of data to be predicted, from a management terminal used by an administrator, and transmits the prediction result to the management terminal.

[0021] The storage unit 12 is a processing unit that stores various data and programs executed by the control unit 20, and is realized by, for example, a memory, a hard disk, etc. The storage unit 12 stores a general-purpose LLM 13 and a difference information DB 14.

[0022] A general-purpose LLM13 is a large-scale language model generated using general training data. Specifically, a general-purpose LLM13 is a general-purpose LLM that is not specialized for a specific domain and is suitable for various natural language processing tasks such as text classification, sentiment analysis, information extraction, text summarization, text generation, and question answering. Generally, a general-purpose LLM is fine-tuned (additional training) to generate an expert LLM specialized for a specific domain.

[0023] The difference information DB14 is a database that stores, in association with each feature, difference information regarding the difference between the weights of the networks of multiple expert LLMs and the weights of the network of a general-purpose LLM. That is, the difference information DB14 stores the difference between the parameters of a trained expert LLM and the parameters of a trained general-purpose LLM.

[0024] Fig. 4 is a diagram illustrating the difference information DB 14. As shown in Fig. 4, the difference information DB 14 stores "features" indicating embedded features obtained from prompts and "differences" indicating difference information between the weights of the expert LLM and the weights of the general-purpose LLM in association with each other.

[0025] In the example of Figure 4, the feature of "math prompt" is associated with "Difference A," which is the difference between the math-specific LLM and the general-purpose LLM that can appropriately predict math prompts. Similarly, the feature of "coding prompt" is associated with "Difference B," which is the difference between the coding-specific LLM and the general-purpose LLM that can appropriately predict coding prompts.

[0026] The "difference" may be the network weight (parameter) itself, or may be compressed information obtained by compressing the difference. The difference information DB 14 may be generated by the pre-processing unit 30 (described later), or may be generated and stored in another device.

[0027] The control unit 20 is a processing unit that controls the entire information processing device 10, and is realized by, for example, a processor. The control unit 20 has a pre-processing unit 30 and a prediction processing unit 40. The pre-processing unit 30 and the prediction processing unit 40 are realized by, for example, electronic circuits included in the processor or processes executed by the processor.

[0028] The pre-processing unit 30 is a processing unit that performs preparations before making a prediction for a new prompt. Specifically, the pre-processing unit 30 generates the difference information DB 14, receives difference information from an external device, and stores it in the difference information DB 14.

[0029] For example, an example of generating differential information will be described using a prompt related to mathematics. FIG. 5 is a diagram illustrating the generation of differential information DB 14. As shown in FIG. 5, pre-processing unit 30 acquires the "mathematics prompt" used in training the mathematics-specialized LLM during the machine learning phase in which additional learning is performed using general-purpose LLM 13. It should be noted that prompts that are highly accurate when predicted by the mathematics-specialized LLM or prompts input to the mathematics-specialized LLM during prediction can also be used, in addition to the prompts used in training the mathematics-specialized LLM. Furthermore, learning data from regular training of the mathematics-specialized LLM can also be used, in addition to additional learning using the general-purpose LLM 13.

[0030] Next, the pre-processing unit 30 inputs the acquired "mathematics prompt" into the general-purpose LLM 13 to calculate features. For example, the pre-processing unit 30 calculates features from the Transformer of the general-purpose LLM 13. Here, the pre-processing unit 30 calculates, as features, "Token embedding," "average of outputs from all layers of the Transformer block for all tokens of the input prompt," "average of outputs from the final layer of the Transformer block for all tokens of the input prompt," "average of outputs from all layers of the Transformer block for the last token of the input prompt," "average of outputs from the final layer of the Transformer block for the last token of the input prompt," etc.

[0031] Then, the pre-processing unit 30 calculates the difference between the network weights of the trained mathematics-specific LLM and the network weights of the general-purpose LLM 13. For example, the pre-processing unit 30 may simply calculate the difference in the network weights, or may compress the calculated weights. Specifically, when generating a mathematics-specific LLM by additional training of the general-purpose LLM 13, the pre-processing unit 30 can calculate the weights of the general-purpose LLM 13 and the mathematics-specific LLM (expert model) and compress the number of parameters using singular value decomposition (SVD) or the like, since the model structures of each LLM are the same.

[0032] Thereafter, the pre-processing unit 30 associates the feature calculated by inputting the "mathematics prompt" into the general-purpose LLM 13 with the weight difference (or compressed information obtained by compressing the difference) and stores them in the difference information DB 14. In this way, the pre-processing unit 30 calculates the feature and difference information of the prompt (learning data) used in training each expert LLM, and stores them in the difference information DB 14.

[0033] Returning to FIG. 3, the prediction processing unit 40 includes a calculation unit 41, an update unit 42, and a prediction unit 43, and is a processing unit that executes prediction for a new prompt that is newly input data to be predicted.

[0034] The calculation unit 41 is a processing unit that calculates a first feature amount (a feature amount of the new prompt) by inputting a new prompt into the general-purpose LLM 13. Specifically, the calculation unit 41 calculates the feature amount when a new prompt is input into the general-purpose LLM 13 that serves as the source of additional learning for each expert LLM.

[0035] 6 is a diagram illustrating the calculation process during prediction. As shown in FIG. 6, when the calculation unit 41 receives a new prompt "When x+y=22 and x=12, what is y?", the calculation unit 41 inputs the received new prompt to the general-purpose LLM 13. Then, the calculation unit 41 calculates the features (embedded features) of the new prompt from the general-purpose LLM 13 using a method similar to the method described in the pre-processing unit 30. Thereafter, the calculation unit 41 outputs the calculated features and the new prompt to the update unit 42.

[0036] The update unit 42 is a processing unit that identifies information for updating the parameters of the general-purpose LLM 13 based on the features of the new prompt calculated by the calculation unit 41, and updates the parameters of the general-purpose LLM 13 using the identified information to be updated. Specifically, the update unit 42 identifies similar features that are similar to the features of the new prompt from among the features stored in the difference information DB 14, and identifies difference information associated with the similar features as information to be updated. Then, the update unit 42 adds the identified difference information to the parameters of the general-purpose LLM 13.

[0037] 7 is a diagram illustrating the update process during prediction. As shown in FIG. 7, the update unit 42 calculates the similarity between the feature of a new prompt and each feature stored in the difference information DB 14 using a common similarity calculation method such as cosine similarity or Euclidean distance. The update unit 42 then identifies the prompt with the highest similarity that is equal to or greater than a threshold value, "What happens when x is multiplied 12 times when x=12?", and obtains the "weight difference A" stored in association with this prompt.

[0038] The update unit 42 then adds the "weight difference A" to the network weight of the general-purpose LLM 13 to generate the updated general-purpose LLM 13. At this time, the update unit 42 does not perform the update if there are no prompts with a similarity greater than or equal to a threshold value or if there are only prompts whose Euclidean distance is greater than a certain value. The update unit 42 outputs the update result to the prediction unit 43.

[0039] The prediction unit 43 is a processing unit that outputs the prediction result obtained by inputting a new prompt into the updated general-purpose LLM 13 with updated weights. That is, the prediction unit 43 predicts the next token using the general-purpose LLM 13 to which the weight difference with the expert LLM trained with similar prompts (data) has been added.

[0040] 8 is a diagram illustrating the prediction process at the time of prediction. As shown in FIG. 8, the prediction unit 43 inputs a new prompt "When x+y=22 and x=12, what is y?" to the updated general-purpose LLM 13 in which the weight difference A has been added to the weight of the general-purpose LLM 13, and obtains the prompt "y is 10," which is the result predicted by the updated general-purpose LLM 13. The prediction unit 43 then transmits the output result prompt "y is 10" to a specified terminal or displays it on a display unit such as a monitor.

[0041] In this way, the prediction unit 43 does not separately invoke an expert LLM, but instead uses the generic LLM 13 as a proxy for the expert LLM to perform predictions of new prompts.

[0042] (Pre-learning process flow) Next, the flow of the pre-training process will be described. Here, as an example, an example will be described in which each expert LLM has already been trained by additional learning of the general-purpose LLM 13, but the pre-processing unit 30 may also train them.

[0043] 9 is a flowchart showing the flow of the pre-learning process. As shown in FIG. 9, the pre-processing unit 30 acquires the learning data used for training the expert LLM (S101), and inputs the learning data into the general-purpose LLM 13 to calculate the feature quantities (S102).

[0044] Next, the pre-processing unit 30 calculates the difference between the weights of the expert LLM and the weights of the general-purpose LLM 13 (S103), and associates the feature with the difference information and registers it in the difference information DB 14 (S104). After that, if the pre-processing unit 30 continues the processing (S105: No), it repeats S101 and subsequent steps, and when the generation of the corresponding difference information is completed (S105: Yes), it ends the processing.

[0045] (Prediction process flow) 10 is a flowchart showing the flow of the prediction process. As shown in FIG. 10, when a new prompt is input (S201: Yes), the prediction processor 40 inputs the new prompt into the general-purpose LLM 13 and calculates a feature (S202). Next, the prediction processor 40 searches the difference information DB 14 for a similar feature that is similar to the calculated feature (S203).

[0046] If a similar feature equal to or greater than the threshold exists (S204: Yes), the prediction processor 40 acquires difference information corresponding to the retrieved similar feature (S205). The prediction processor 40 then adds the difference information to the general-purpose LLM 13 (S206), inputs a new prompt to the general-purpose LLM 13 after the addition, and predicts the next token (S207). The prediction processor 40 then outputs the prediction result (S208).

[0047] On the other hand, if there are no similar features equal to or greater than the threshold (S204: No), the prediction processing unit 40 inputs a new prompt to the general-purpose LLM 13 that has not performed addition, and predicts the next token (S209). Thereafter, the prediction processing unit 40 outputs the prediction result (S210).

[0048] (effect) As described above, the information processing device 10 uses the embedded features of the general-purpose LLM 13 itself to calculate what data the embedded input features are similar to, and predicts the next token by changing the weights of the general-purpose LLM 13 so that they approach the weights of the specialized LLM associated with similar data.

[0049] Therefore, the information processing device 10 does not need to prepare an expert LLM or a classifier that performs routing. In other words, the information processing device 10 can substitute a classifier and multiple expert LLMs with a single general-purpose LLM 13, thereby reducing the amount of spatial calculation.

[0050] Furthermore, the information processing device 10 can reduce the memory capacity of the expert LLM and the classifier. Furthermore, since the information processing device 10 can use a large memory capacity for prediction processing, it is possible to suppress processing delays and reduce the processing load on the computer.

[0051] Furthermore, the information processing device 10 generates difference information including features and differences from the training data used for training the expert LLM. Therefore, the information processing device 10 can control the general-purpose LLM 13 to which the difference information is added in the same way as an actual expert LLM, thereby suppressing a decrease in prediction accuracy.

[0052] Furthermore, the information processing device 10 can store the differential information in a compressed state in the DB and add it to the weight in the compressed state, thereby reducing the memory capacity more than when the differential information is stored as is. [Example]

[0053] Although the embodiments of the present invention have been described above, the present invention may be embodied in various different forms other than the above-described embodiments.

[0054] (Numbers, etc.) The machine learning model, the number of LLMs, the training data, the prompts, etc. used in the above examples are merely examples and can be changed as desired. Furthermore, the process flow described in each flowchart can also be changed as appropriate within a consistent range.

[0055] (multiple weighted differences) For example, in the above example, the information processing device 10 acquires the weighted difference of the most similar feature from the difference information DB 14 and adds it to the general-purpose LLM 13, but this is not limiting. For example, the information processing device 10 can update (extend or complement) the general-purpose LLM 13 using multiple pieces of difference information.

[0056] FIG. 11 is a diagram illustrating an example of processing using multiple weighted differences. As shown in FIG. 11, the information processing device 10 refers to the difference information DB 14 and searches for feature quantities similar to the input prompt. Here, the information processing device 10 searches for three pieces of difference information, "weighted difference 1," "weighted difference 2," and "weighted difference 3," whose similarity is equal to or greater than a threshold. In this case, the information processing device 10 can add the average values ​​and variances of "weighted difference 1," "weighted difference 2," and "weighted difference 3" to the general-purpose LLM 13. Note that the information processing device 10 may acquire any number of weighted differences in descending order of similarity.

[0057] As a result, the information processing device 10 can suppress a decrease in prediction accuracy as much as possible even when there is a large variation in weight information due to the influence of learning data or the like.

[0058] (Another example of prediction processing) For example, in the process described in the first embodiment, the next prompt may be predicted instead of the final answer in one prediction. In such a case, the process up to obtaining the final answer will be described.

[0059] FIG. 12 is a diagram illustrating another example of prediction processing using an updated general-purpose LLM 13. As shown in FIG. 12, the information processing device 10 inputs a new prompt into the "general-purpose LLM 13 to which the weighted difference A has been added" generated using the method of the first embodiment, and obtains "prompt X1" as a prediction result. Next, the information processing device 10 inputs "prompt X1" into the "general-purpose LLM 13 to which the weighted difference A has been added," and obtains "prompt X2" as a prediction result. Thereafter, the information processing device 10 repeats the prediction processing of inputting the prediction result into the "general-purpose LLM 13 to which the weighted difference A has been added" until an answer to the new prompt is obtained.

[0060] In this way, the information processing device 10 can obtain a final prediction result (token) by repeatedly using the general-purpose LLM 13 that has been updated (extended) once, thereby realizing a short prediction processing time.

[0061] The information processing device 10 can also perform an update process each time a prompt is acquired as a prediction result. Fig. 13 is a diagram illustrating an example of updating the general-purpose LLM 13. As shown in Fig. 13, when a new prompt is input, the information processing device 10 identifies a difference A based on the similarity using the method of the first embodiment, and inputs the new prompt into the general-purpose LLM 13 to which the difference A has been added, thereby acquiring the prediction result "prompt X1."

[0062] Next, similar to the method of the first embodiment, the information processing device 10 inputs the prediction result "prompt X1" into the general-purpose LLM 13 to calculate a feature amount, and identifies the difference B between the feature amount and a feature amount similar to the calculated feature amount. After that, the information processing device 10 inputs "prompt X1" into the general-purpose LLM 13 to which the difference B has been added, and obtains the prediction result "prompt X2."

[0063] In this way, the information processing device 10 can repeat updating and prediction of the general-purpose LLM 13 in response to input prompts until a final prediction result is obtained, thereby improving prediction accuracy.

[0064] (system) The information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings may be changed arbitrarily unless otherwise specified.

[0065] Furthermore, the specific form of distribution or integration of the components of each device is not limited to that shown in the figure. For example, the pre-processing unit 30 and the prediction processing unit 40 may be integrated. That is, all or some of the components may be functionally or physically distributed or integrated in any unit depending on various loads, usage conditions, etc. Furthermore, all or any part of the processing functions of each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0066] Furthermore, all or any part of the processing functions performed by each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0067] (Hardware) Fig. 14 is a diagram illustrating an example of a hardware configuration. As shown in Fig. 14, an information processing device 10 includes a communication device 10a, a hard disk drive (HDD) 10b, a memory 10c, and a processor 10d. The components shown in Fig. 14 are connected to each other via a bus or the like.

[0068] The communication device 10a is a network interface card or the like, and communicates with other devices. The HDD 10b stores programs and DBs that operate the functions shown in FIG.

[0069] The processor 10d reads out a program that executes the same processes as the respective processing units shown in FIG. 3 from the HDD 10b or the like and loads it into the memory 10c, thereby operating a process that executes the respective functions described in FIG. 3 or the like. For example, this process executes the same functions as the respective processing units of the information processing device 10. Specifically, the processor 10d reads out a program having the same functions as the pre-processing unit 30, the prediction processing unit 40, or the like from the HDD 10b or the like. Then, the processor 10d executes a process that executes the same processes as the pre-processing unit 30, the prediction processing unit 40, or the like.

[0070] In this way, the information processing device 10 operates as an information processing device that executes a prediction method by reading and executing a program. The information processing device 10 can also realize functions similar to those of the above-described embodiment by reading the program from a recording medium using a medium reading device and executing the read program. Note that the program in these other embodiments is not limited to being executed by the information processing device 10. For example, the above-described embodiment may also be applied in the same way to cases where another computer or server executes the program, or where these execute the program in cooperation with each other.

[0071] This program may be distributed via a network such as the Internet. Alternatively, this program may be recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), or a digital versatile disk (DVD), and may be read out from the recording medium and executed by a computer. [Explanation of symbols]

[0072] 10. Information processing equipment 11 Communications Department 12 Storage section 13 General LLM 14 Difference information DB 20 Control Unit 30 Pre-processing section 40 Prediction processing unit 41 Calculation section 42 Update section 43 Prediction Department

Claims

1. On the computer, inputting first data into a first machine learning model to calculate a first feature amount; Identifying information for updating parameters of the first machine learning model based on the first feature amount; updating parameters of the first machine learning model using the updating information; inputting the first data into the first machine learning model whose parameters have been updated, and outputting the obtained prediction result; A prediction program that executes a process.

2. inputting a plurality of second data items suitable for each of a plurality of second machine learning models into the first machine learning model to calculate each feature amount; calculating a difference between a parameter of each of the plurality of second machine learning models and a parameter of the first machine learning model; generating a database in which each of the feature amounts is associated with difference information relating to the difference in the parameters; 2. The prediction program according to claim 1, wherein the program causes a computer to execute processing.

3. The calculation process includes: compressing the difference between the parameters of each of the plurality of second machine learning models and the parameters of the first machine learning model; Compressed information obtained by compressing the difference is associated with each of the feature amounts as the difference information.

3. The prediction program according to claim 2.

4. The identifying process includes: referring to the database, and identifying, as the information to be updated, the difference information associated with a similar feature amount similar to the first feature amount among the feature amounts; The updating process includes: adding the difference information to parameters of the first machine learning model; The output process includes: inputting the first data into the first machine learning model to which the difference information has been added, and outputting the obtained prediction result; 4. The prediction program according to claim 2 or 3.

5. The identifying process includes: identifying, as the information to be updated, each piece of difference information associated with a plurality of similar feature amounts similar to the first feature amount; The updating process includes: adding an average value of each piece of difference information to a parameter of the first machine learning model; 5. The prediction program according to claim 4.

6. each of the plurality of second machine learning models is a model generated by additionally training the first machine learning model that has already been trained; the plurality of second data are data used for additional learning of any of the plurality of second machine learning models; 3. The prediction program according to claim 2.

7. the first machine learning model is a large-scale language model; the first data is a prompt; The output process includes: repeating inputting the output result of the first machine learning model whose parameters have been updated into the first machine learning model whose parameters have been updated until an answer to the prompt is obtained; The prediction program according to claim 1 .

8. the first machine learning model is a large-scale language model; the first data is a prompt; The calculation process includes: inputting second data, which is a prediction result of the first machine learning model with the updated parameters, into the first machine learning model before the update to calculate a third feature amount; The identifying process includes: Identifying second information for updating parameters of the first machine learning model based on the third feature amount; The updating process includes: updating parameters of the first machine learning model using the second information; The output process includes: inputting the second data into the first machine learning model updated by the second information, and outputting the obtained prediction result; The prediction program according to claim 1 .

9. The computer inputting first data into a first machine learning model to calculate a first feature amount; Identifying information for updating parameters of the first machine learning model based on the first feature amount; updating parameters of the first machine learning model using the updating information; inputting the first data into the first machine learning model whose parameters have been updated, and outputting the obtained prediction result; A prediction method comprising:

10. inputting first data into a first machine learning model to calculate a first feature amount; Identifying information for updating parameters of the first machine learning model based on the first feature amount; updating parameters of the first machine learning model using the updating information; inputting the first data into the first machine learning model whose parameters have been updated, and outputting the obtained prediction result; An information processing device comprising a control unit.

Images