Information processing apparatus, information processing method, and program

Abstract

To provide a technique that can suitably prevent overtraining.SOLUTION: An information processing apparatus comprises: an acquisition unit that acquires target data and the number of a plurality of target models; a regression coefficient calculation unit that refers to the target data and a ratio parameter defining the ratio between the plurality of target models, and calculates a sparse regression coefficient for each of the plurality of target models; a ratio parameter calculation unit that refers to the target data and the regression coefficients to calculate the ratio parameter; and an output unit that outputs the regression coefficients calculated by the regression coefficient calculation means and the ratio parameter calculated by the ratio parameter calculation means.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Among various optimization methods and control methods, methods using adjustment parameters called scheduling parameters are known. For example, Patent Document 1 discloses a technology related to gain schedule control using scheduling parameters. In control methods and optimization methods using multiple models and scheduling parameters, the scheduling parameters are sometimes used as parameters that define the ratios between these multiple models. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-030414 Summary of the Invention [Problem to be solved by the invention]

[0004] It is generally known that in optimization methods, a problem called overfitting can occur in the parameter optimization (learning) process. However, when multiple models and scheduling parameters (ratio parameters) are used, it is not known how to suppress overfitting, and even if the technology of Patent Document 1 is used, it is difficult to suppress overfitting in such a case.

[0005] The present disclosure has been made in consideration of the above-mentioned problems, and one exemplary purpose thereof is to provide a technology that can suitably suppress overlearning in a configuration that uses scheduling parameters (ratio parameters). [Means for solving the problem]

[0006] An information processing device according to an exemplary aspect of the present disclosure includes an acquisition means for acquiring target data and the number of models of a plurality of target models, a regression coefficient calculation means for calculating a sparse regression coefficient for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models, a ratio parameter calculation means for calculating the ratio parameter by referring to the target data and the regression coefficient, and an output means for outputting the regression coefficient calculated by the regression coefficient calculation means and the ratio parameter calculated by the ratio parameter calculation means.

[0007] An information processing device according to an exemplary aspect of the present disclosure includes an acquisition means for acquiring inference data and information relating to the type of the inference data, a prediction means for deriving a prediction result by applying to the inference data a regression coefficient for each of a plurality of target models and a ratio parameter that specifies a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data, and an output means for outputting the prediction result by the prediction means, wherein the regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process that includes a regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameter and training data, and a ratio parameter calculation process that calculates the ratio parameter by referring to the training data and the regression coefficients.

[0008] An information processing method according to an exemplary aspect of the present disclosure includes a computer acquiring target data and the number of models of a plurality of target models, calculating sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models, calculating the ratio parameter by referring to the target data and the regression coefficients, and outputting the regression coefficients and the ratio parameter.

[0009] An information processing method according to an exemplary aspect of the present disclosure includes, by a computer, acquiring inference data and information related to the type of the inference data, deriving a prediction result by applying to the inference data regression coefficients for each of a plurality of target models and ratio parameters that specify a ratio of the plurality of target models, the ratio parameters being determined according to the type of the inference data, and outputting the prediction result, wherein the regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process that includes a regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and training data, and a ratio parameter calculation process that calculates the ratio parameters by referring to the training data and the regression coefficients.

[0010] The information processing device according to each aspect of the present invention may be realized by a computer, in which case a program that causes the computer to operate as each part (software element) of the information processing device to realize the information processing device also falls within the scope of the present invention. [Effects of the Invention]

[0011] According to an exemplary aspect of the present disclosure, in a configuration using a scheduling parameter (ratio parameter), an exemplary effect is achieved in which overlearning can be suitably suppressed. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 2] FIG. 1 is a flow diagram showing the flow of an information processing method according to the present disclosure. [Figure 3] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 4] FIG. 1 is a flow diagram showing the flow of an information processing method according to the present disclosure. [Figure 5] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 6] FIG. 10 is a diagram schematically illustrating the output of each endpoint model in the LPV model according to the present disclosure and the interior division ratio parameters by which each output is multiplied. [Figure 7] FIG. 10 is a diagram illustrating an example of a processing flow in an information processing device according to the present disclosure. [Figure 8] 10 shows an example of a graph displayed by an output unit via an input / output unit according to the present disclosure. [Figure 9] 10 shows an example of a graph displayed by an output unit via an input / output unit according to the present disclosure. [Figure 10] 10 shows an example of a graph displayed by an output unit via an input / output unit according to the present disclosure. [Figure 11] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 12] 1 is a block diagram illustrating a configuration of an information processing device according to the present disclosure. [Figure 13] FIG. 10 is a diagram illustrating an example of a processing flow in an information processing device according to the present disclosure. [Figure 14] FIG. 1 is a block diagram illustrating a configuration of a computer that functions as an information processing device according to the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] The following are examples of embodiments of the present invention. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technical means employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, embodiments obtained by appropriately omitting some of the technical means employed in the exemplary embodiments shown below may also be included in the scope of the present invention. Furthermore, the effects mentioned in the exemplary embodiments shown below are examples of effects expected in the exemplary embodiments, and do not define the scope of the present invention. In other words, embodiments that do not exhibit the effects mentioned in the exemplary embodiments shown below may also be included in the scope of the present invention.

[0014] First Exemplary Embodiment A first exemplary embodiment, which is one example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is the basic form of each exemplary embodiment described later. Note that the scope of application of each technical means employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure to the extent that no particular technical obstacles arise. Furthermore, each technical means shown in the drawings referred to in describing this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure to the extent that no particular technical obstacles arise.

[0015] (Configuration of information processing device 1) The configuration of the information processing device 1 will be described with reference to Fig. 1. Fig. 1 is a block diagram showing the configuration of the information processing device 1. As shown in Fig. 1, the information processing device 1 includes an acquisition unit 11, a regression coefficient calculation unit 12, a ratio parameter calculation unit 14, and an output unit 15. In this exemplary embodiment, the acquisition unit 11, the regression coefficient calculation unit 12, the ratio parameter calculation unit 14, and the output unit 15 respectively realize an acquisition means, a regression coefficient calculation means, a ratio parameter calculation means, and an output means.

[0016] (Acquisition part 11) The acquisition unit 11 acquires target data and information regarding the number of models of a plurality of target models. The acquisition unit 11 supplies the acquired target data to the regression coefficient calculation unit 12 and the ratio parameter calculation unit 14. The acquisition unit 11 also supplies the acquired number of models of the plurality of target models to the regression coefficient calculation unit 12.

[0017] (Regression coefficient calculation part 12) The regression coefficient calculation unit 12 calculates sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines the ratio of the plurality of target models. Here, the specific method of calculating the sparse regression coefficients by the regression coefficient calculation unit 12 does not limit this exemplary embodiment, but as an example, calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the sparse regression coefficients by sparse linear regression processing that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. The regression coefficient calculation unit 12 supplies the calculated regression coefficients to the ratio parameter calculation unit 14 and the output unit 15.

[0018] (Ratio parameter calculation part 14) The ratio parameter calculation unit 14 calculates the ratio parameters by referring to the target data and the regression, and supplies the calculated ratio parameters to the output unit 15.

[0019] (Output section 15) The output unit 15 outputs the regression coefficients calculated by the regression coefficient calculation unit 12 and the ratio parameters calculated by the ratio parameter calculation unit 14. The regression coefficients and interior division ratio parameters output by the output unit 15 are, for example, stored in a storage unit (not shown) or provided to a device external to the information processing device 1 by an input / output unit (not shown).

[0020] (Effects of information processing device 1) As described above, in the information processing device 1, Obtain the target data and the number of models for multiple target models, calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the ratio parameter by referring to the target data and the regression coefficients; Outputting the regression coefficients and the ratio parameters The following configuration is adopted.

[0021] According to the above configuration, sparse regression coefficients for each of the plurality of target models are calculated by referring to the ratio parameters and the target data, so that it is possible to perform optimization processing (learning processing, update processing) for the ratio parameters while suppressing overlearning. Therefore, according to the above configuration, it is possible to suitably suppress overlearning in a configuration that uses scheduling parameters (ratio parameters).

[0022] (Flow of information processing method S1) The flow of the information processing method S1 will be described with reference to Fig. 2. Fig. 2 is a flow diagram showing the flow of the information processing method S1. As shown in Fig. 2, the information processing method S1 includes an acquisition process S11, a regression coefficient calculation process S12, a ratio parameter calculation process S14, and an output process S15.

[0023] (Step S11) In step S11, the acquisition unit 11 acquires target data and information regarding the number of models of multiple target models. The acquisition unit 11 supplies the acquired target data to the regression coefficient calculation unit 12 and the ratio parameter calculation unit 14. The acquisition unit 11 also supplies the acquired number of models of the multiple target models to the regression coefficient calculation unit 12.

[0024] (Step S12) In step S12, the regression coefficient calculation unit 12 calculates sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines the ratio of the plurality of target models. The regression coefficient calculation unit 12 supplies the calculated regression coefficients to the ratio parameter calculation unit 14 and the output unit 15. Note that the specific processing by the regression coefficient calculation unit 12 has been described above, and therefore will not be described here.

[0025] (Step S14) Subsequently, in step S13, the ratio parameter calculation unit 14 calculates ratio parameters by referring to the target data and the regression. The ratio parameter calculation unit 14 supplies the calculated ratio parameters to the output unit 15.

[0026] (Step S15) Subsequently, in step S15, the output unit 15 outputs the regression coefficients calculated by the regression coefficient calculation unit 12 and the ratio parameters calculated by the ratio parameter calculation unit 14. The regression coefficients and the interior division ratio parameters output by the output unit 15 are, for example, stored in a storage unit (not shown) or provided to a device external to the information processing device 1 by an input / output unit (not shown).

[0027] (Effect of information processing method S1) As described above, in the information processing method S1, Obtain the target data and the number of models for multiple target models, calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the ratio parameter by referring to the target data and the regression coefficients; Outputting the regression coefficients and the ratio parameters According to the above configuration, the same effects as those of the information processing device 1 are achieved.

[0028] (Configuration of information processing device 2) The configuration of the information processing device 2 will be described with reference to Fig. 3. Fig. 3 is a block diagram showing the configuration of the information processing device 2. As shown in Fig. 3, the information processing device 2 includes an acquisition unit 21, a prediction unit 22, and an output unit 23. The acquisition unit 21, the prediction unit 22, and the output unit 23 each include In this exemplary embodiment, an obtaining means, a predicting means, and an outputting means are provided.

[0029] (Acquisition part 21) The acquisition unit 21 acquires data for inference and information on the type of the data for inference. The acquisition unit 21 supplies the acquired data for inference to the prediction unit 22.

[0030] (Prediction Section 22) The prediction unit 22 derives a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models and is determined according to the type of the inference data to the inference data acquired by the acquisition unit 21. Here, the regression coefficient and the ratio parameter are A regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process including the above. Furthermore, the ratio parameters determined according to the type of inference data refer, for example, to ratio parameters learned (updated) by a learning process using learning data (target data) associated with a type that is the same as or similar to the type of the inference data, but this does not limit the present exemplary embodiment. For example, the regression coefficients and the ratio parameters are regression coefficients calculated by the information processing device 1 according to the present exemplary embodiment. The prediction unit 22 supplies the prediction result to the output unit 23.

[0031] (output unit 23) The output unit 23 outputs the prediction result by the prediction unit 22. As an example, the output unit 23 visually presents the prediction result to the user via a display means (not shown).

[0032] (Effects of information processing device 2) As described above, in the information processing device 2, Acquire inference data and information regarding the type of the inference data, deriving a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models to the inference data, the ratio parameter being determined according to the type of the inference data; Outputting the prediction result by the prediction means In this configuration, The regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process including a regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and learning data, and a ratio parameter calculation process that calculates the ratio parameters by referring to the learning data and the regression coefficients. The following configuration is adopted.

[0033] According to the above configuration, sparse regression coefficients for each of the plurality of target models are calculated by referring to the ratio parameters and the target data, and therefore, optimization processing (learning processing, update processing) for the ratio parameters is performed while suppressing overfitting. Therefore, according to the above configuration, overfitting is preferably suppressed in a configuration using scheduling parameters (ratio parameters). Furthermore, prediction can be preferably performed using regression coefficients in which overfitting is suppressed.

[0034] (Flow of information processing method S2) The flow of the information processing method S2 will be described with reference to Fig. 4. Fig. 4 is a flow diagram showing the flow of the information processing method S2. As shown in Fig. 4, the information processing method S2 includes an acquisition process S21, a prediction process S22, and an output process S23.

[0035] (Acquisition process S21) In the acquisition process S21, the acquisition unit 21 acquires data for inference and information on the type of the data for inference. The acquisition unit 21 supplies the acquired data for inference to the prediction unit 22.

[0036] (Prediction process S22) In the prediction process S22, the prediction unit 22 derives a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models and is determined according to the type of the inference data to the inference data acquired by the acquisition unit 21. Here, the regression coefficient and the ratio parameter are A regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The prediction unit 22 outputs the prediction result to the output unit 23. ... (Output process S23) In the output process S23, the output unit 23 outputs the prediction result by the prediction unit 22. As an example, the output unit 23 visually presents the prediction result to the user via a display means (not shown).

[0037] (Effect of information processing method S2) As described above, in the information processing method S2, Acquire inference data and information regarding the type of the inference data, deriving a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models to the inference data, the ratio parameter being determined according to the type of the inference data; Outputting the prediction result by the prediction means In this configuration, The regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process including a regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and learning data, and a ratio parameter calculation process that calculates the ratio parameters by referring to the learning data and the regression coefficients. The above configuration provides the same effects as the information processing device 2 described above.

[0038] Second Exemplary Embodiment A second exemplary embodiment, which is one example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same functions as those described in the above exemplary embodiment will be assigned the same reference numerals, and their description will be omitted as appropriate. The scope of application of each technical means employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. Furthermore, each technical means shown in each drawing referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. (Overview of information processing device 1A) First, an overview of an information processing device 1A according to this exemplary embodiment will be described. Obtain the target data and the number of models for multiple target models, calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the ratio parameter by referring to the target data and the regression coefficients; Outputting the regression coefficients and the ratio parameters According to the above configuration, sparse regression coefficients for each of the plurality of target models are calculated by referring to the ratio parameters and the target data, so that it is possible to perform optimization processing (learning processing, update processing) for the ratio parameters while suppressing overlearning. Therefore, according to the above configuration, it is possible to suitably suppress overlearning in a configuration that uses scheduling parameters (ratio parameters).

[0039] In the above configuration, the specific algorithm for calculating the "regression coefficient" is not particularly limited, and may be the least squares method or another algorithm. In this exemplary embodiment, a process including, as examples, a Linear Parameter-Varying (LPV) model and an L2PV (Latent Linear Parameter-Varying) model, which are being studied by the present inventors, will be described, but these examples do not limit the present exemplary embodiment.

[0040] The positioning of the algorithm of the processing by the information processing device 1A according to this exemplary embodiment will be described. The present inventors have been studying a Linear Parameter-Varying (LPV) model as a modeling of a system with fluctuations. In the LPV model, as an example, an internal state quantity (internal state variable) x k , and the output state quantity (output state variable) y k is updated and calculated using the following formulas (1A) and (1B).

number

[0041] Figure 6 shows the output of each end model in the LPV model (1st SS model to 5th SS model in Figure 6) and the internal division ratio parameter μ (i) k As shown in Fig. 6, the output of multiple endpoint models at the k-th step is (A (i) x k +B (i) u k ) (i=1~5) For each of the internal division ratio parameters μ (i) k (i=1~5) is multiplied, and x at the k+1th step is k+1 is calculated.

[0042] While such an LPV model is suitable for modeling systems with fluctuations, the internal ratio parameter μ (i) k However, there was a problem that it was difficult to apply to systems where the value of is unknown.

[0043] The present inventors The above internal division ratio parameter μ (i) k The hidden variable (posterior probability) z k Treat it as - Applying hidden variable model learning methods in machine learning, The above internal division ratio parameter μ (i) k the hidden variable z k Calculate as the expected value of By doing so, the inventors have found that it is possible to realize the learning of an LPV model even if the interior ratio parameter μ (i) kThe L2PV model (Latent Linear Parameter-Varying model) is defined by the following equations (2A) to (2C), which introduce the following as hidden variables:

number

number

[0044] Each process performed by the information processing device 1A described below is based on the above-mentioned formulation and is a process based on the unique viewpoint of the inventor.

[0045] (Configuration of information processing device 1A) The configuration of the information processing device 1A will be described with reference to Fig. 5. Fig. 5 is a block diagram showing the configuration of the information processing device 1A. As shown in Fig. 5, the information processing device 1A includes a control unit 10A, a storage unit 15A, a communication unit 16A, and an input / output unit 17A.

[0046] (Storage section 15A) First, various data (information) stored in the storage unit 15A will be described. Data referenced by the control unit 10A is stored in the storage unit 15A. Examples of the storage unit 15A include, but are not limited to, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0047] Examples of data stored in memory unit 15A include, but are not limited to, target data TD, internal division ratio parameters RP, regression coefficients RC, distribution information DI, learning results LR, inference data PD, and prediction results PR, as shown in FIG. 5.

[0048] The target data TD is data used in the learning process in the information processing device 1A. The target data TD is a state variable (~x k ) and state variables (~y k ) is expressed as the following equation (4).

number

number

[0049] Also, the internal ratio parameter μ for a certain model jk (j) is the N-dimensional target data x k It can also be expressed as the components of an N-dimensional vector having components corresponding to each of the j-th internal division ratio parameters μ k (j) is the N-dimensional target data x k (k=1 to N) (j) , μ2 (j) , , μ N (j) ) are the components of an N-dimensional vector with

[0050] The regression coefficient RC is a coefficient in the L2PV regression model. The regression coefficient RC is expressed as the following equation (6).

number

[0051] The correspondence information CI is information about the correspondence between the target data TD and a plurality of target models. Examples of the information about the correspondence include: -Information indicating what kind of target data TD each of the multiple target models has been trained (will be trained) with is included. For example, the correspondence relationship information CI relating to the target models a, b, and c and the target data TD1, TD2, and TD3 is as follows: Target model a: Target data TD1, TD2 Target model b: Target data TD3 Target model c: Target data TD1, TD2, TD3 Here, the first line in the above example indicates that the target model a has been trained (will be trained) using the target data TD1 and TD2. The same applies to the other lines.

[0052] In other words, the correspondence information CI is Information indicating whether each of the multiple target models has used (will use) each of the multiple target data TD as training data As an example, the correspondence relationship information CI relating to the target models a, b, and c and the target data TD1, TD2, and TD3 may be expressed as follows: Target model a: Target data TD1 (○), TD2 (○), TD3 (×) Target model b: Target data TD1(×), TD2(×), TD3(○) Target model c: Target data TD1(○), TD2(○), TD3(○) Here, ○ indicates that it was used (will be used) for learning, and × indicates that it was not used (will not be used) for learning.

[0053] In the above example, the correspondence relationship information CI may include information about the circumstances under which each piece of target data was acquired. Target data TD1 (e.g., sales volume): Situation A (e.g., sunny weather) Target data TD2 (e.g., sales volume): Situation B (e.g., cloudy weather) Target data TD3 (e.g., sales volume): Situation C (e.g., the weather is rainy) In addition, when the correspondence information CI includes the correspondence between each target data and the situation in which the target data was acquired in this way, the correspondence information CI may be expressed as including the relationship between each target model and information related to the target model. For example, in the above example, The correspondence information CI is Target model a: Situation A (e.g., sunny weather), Situation B (e.g., cloudy weather) Target model b: Situation C (e.g., rainy weather) Target model c: Situations A, B, and C The correspondence relationship information CI may include information indicating a correspondence such as the following. By including the above information in the correspondence relationship information CI, the information processing device 1A can identify what situation each target model is associated with, in other words, what situation each target model is preferably used in. Furthermore, the information processing device 1A can also present to the user what situation each target model is preferably used in.

[0054] The information about situations such as situations A, B, and C may be expressed as information about types. In other words, the correspondence relationship information CI may include information indicating which type at least one of the target model and the target data is associated with.

[0055] In addition, the correspondence information CI is Information indicating whether a certain target data TD is similar to other target data TD Here, the target data TD and other target data TD being similar refers to the case where the types of the target data and other target data TD are similar. For example, if the target data TD is data on the number of sales acquired in situation A (e.g., sunny weather), then the target data TD', which is data on the number of sales in situation A' (e.g., sunny weather) similar to situation A, is data similar to the target data TD.

[0056] The distribution information DI includes a covariance matrix Φ of the prior distribution of the latent variables, a covariance parameter η of the prior distribution of the latent variables, and a covariance parameter Ψ of the posterior distribution of the latent variables.

[0057] Hidden variable z k The prior distribution p(z k ) is expressed as the following equation (7).

number

number

number

[0058] As will be described later, in the processing by the information processing device 1A, the hidden variable z k The posterior distribution p(z k |~y k ,~x k , W, η) is expressed under the constraints (constraints) of the following equation (10).

number

[0059] The learning result LR is data that is output by the output unit 15, which will be described later. The learning result LR includes the calculated internal ratio parameter RP and the calculated regression coefficient RC.

[0060] The inference data PD is an internal state quantity input to the L2PV regression model. The L2PV regression model makes predictions by using the inference data PD as input and applying the calculated regression coefficients RC to each of multiple target models.

[0061] The predicted result PR is a predicted result obtained by the L2PV regression model. Examples of the predicted result PR will be described later.

[0062] (Communication unit 16A) Communication unit 16A is an interface that transmits and receives data via a network. Examples of communication unit 16A include, but are not limited to, communication chips for various communication standards such as Ethernet (registered trademark), Wi-Fi (Wireless Fidelity (registered trademark), and wireless communication standards for mobile data communication networks, and USB-compliant connectors.

[0063] (Input / output section 17A) The input / output unit 17A is an interface that receives input of data and outputs data. Examples of the input / output unit 17A include, but are not limited to, a microphone, a camera, a gaze input device, a keyboard, a touchpad, a speaker, and a liquid crystal display.

[0064] (Control unit 10A) The control unit 10A controls each component included in the information processing device 1A. As shown in FIG. 5 , the control unit 10A also includes an acquisition unit 11, a regression coefficient calculation unit 12, a covariance calculation unit 13, an interior ratio parameter calculation unit 14, an output unit 15, an initial value determination unit 16, a convergence determination unit 17, and a prediction unit 18. In this exemplary embodiment, the acquisition unit 11, the regression coefficient calculation unit 12, the covariance calculation unit 13, the interior ratio parameter calculation unit 14, the output unit 15, the initial value determination unit 16, the convergence determination unit 17, and the prediction unit 18 function as an acquisition means, a regression coefficient calculation means, a covariance calculation means, an interior ratio parameter calculation means, an output means, an initial value determination means, a convergence determination means, and a prediction means, respectively. Specific examples of the processing performed by each unit will be described later with reference to different drawings.

[0065] The acquisition unit 11 acquires data via the communication unit 16A or the input / output unit 17A. Examples of the data acquired by the acquisition unit 11 include target data TD, the number of models of a plurality of target models, and hidden variables z kAnother example of data acquired by the acquisition unit 11 is inference data PD. The acquisition unit 11 stores the acquired data in the storage unit 15A.

[0066] The regression coefficient calculation unit 12 calculates sparse regression coefficients RC for each of the plurality of target models by referring to the target data TD and an interior division ratio parameter RP that defines the interior division ratio of the plurality of target models. calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the sparse regression coefficients by sparse linear regression processing that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. Here, as the sparse linear regression process, for example, Lasso regression may be performed. More specific processing by the regression coefficient calculation unit 12 will be described later.

[0067] As an example, the interior ratio parameter RP referred to by the regression coefficient calculation unit 12 is the initial value of the interior ratio parameter RP determined by an initial value determination unit 16 (described later). As another example, the interior ratio parameter RP referred to by the regression coefficient calculation unit 12 is the interior ratio parameter RP calculated by an interior ratio parameter calculation unit 14 (described later). The regression coefficient calculation unit 12 stores the calculated regression coefficient RC in memory unit 15A.

[0068] The covariance calculation unit 13 calculates the target data TD, the interior ratio parameter RP, the regression coefficient RC, and the hidden variable z k By referring to the covariance matrix Φ of the prior distribution of the hidden variable z k The covariance parameter η of the prior distribution of and the hidden variable z k The covariance calculation unit 13 calculates the covariance matrix Ψ of the posterior distribution of the calculated hidden variable z k The covariance parameter η of the prior distribution of and the hidden variable z k and the covariance matrix Ψ of the posterior distribution of the above are stored in the storage unit 15A as distribution information DI.

[0069] The internal ratio parameter calculation unit 14 calculates the target data TD, the regression coefficients RC, and the hidden variables z k The internal ratio parameter calculation unit 14 calculates the internal ratio parameter RP by referring to the covariance matrix Ψ of the posterior distribution of the internal ratio parameter RP. The internal ratio parameter calculation unit 14 stores the calculated internal ratio parameter RP in the storage unit 15A.

[0070] The output unit 15 outputs the regression coefficient RC and the interior ratio parameter RP (learning result LR) calculated by the interior ratio parameter calculation unit 14. As one example, the output unit 15 outputs an image including the regression coefficient RC and the interior ratio parameter RP calculated by the interior ratio parameter calculation unit 14 to the input / output unit 17A. As another example, the output unit 15 outputs the regression coefficient RC and the interior ratio parameter RP calculated by the interior ratio parameter calculation unit 14 when the convergence determination unit 17 described later determines that the calculation related to the interior ratio parameter RP has converged.

[0071] Initial value determination unit 16 determines the initial value of interior ratio parameter RP referenced by regression coefficient calculation unit 12. Initial value determination unit 16 stores the determined initial value of interior ratio parameter RP in storage unit 15A.

[0072] The convergence determination unit 17 determines whether or not the calculation related to the interior ratio parameter RP has converged. The convergence determination unit 17 supplies the determination result to the output unit 15.

[0073] The prediction unit 18 derives multiple prediction results PR by applying the regression coefficients RC for each of the multiple target models to the inference data PD. The prediction unit 18 stores the derived prediction results PR in the storage unit 15A. Therefore, the prediction unit 18 can derive multiple prediction results PR for each regression coefficient RC.

[0074] (Example of processing flow in information processing device 1A) 7 is a diagram showing an example of a processing flow in the information processing device 1A according to this exemplary embodiment. Note that the processing example described below can also be regarded as a variational Bayes EM algorithm, but this does not limit this exemplary embodiment. Furthermore, the processing example described below can be regarded as processing for updating each parameter so as to maximize a variational lower bound (VLB) J obtained by the following equation (11).

number

number

[0075] In step S11, the acquisition unit 11 further acquires a parameter m indicating the number of models among the plurality of target models. Here, the number of models m is determined by the internal division ratio parameter vector μ k (i) It can also be expressed as the number of

[0076] In step S11, the acquisition unit 11 calculates the hidden variable z k As an example, the acquisition unit 11 acquires information about the prior distribution of the hidden variable z k The prior distribution p(z k ) covariance matrix Φ. The acquisition unit 11 also acquires the hidden variable z k The prior distribution p(z k ) may further be obtained.

[0077] (Step S16: Initial value determination process) Next, in step S16, the initial value determination unit 16 determines the initial value of the interior ratio parameter RP to be referenced in the regression coefficient calculation process S12, which will be described later. As an example, the initial value determination unit 16 determines the initial value of the interior ratio parameter RP as a random value. By the initial value determination unit 16 determining the initial value of the interior ratio parameter RP in this way, the regression coefficient RC can be suitably calculated in the regression coefficient calculation process S12, which will be described later. The details of the interior ratio parameter RP have been explained above, so a detailed explanation will be omitted here.

[0078] (Step S12: Regression coefficient calculation process) Subsequently, in step S12, the regression coefficient calculation unit 12 calculates sparse regression coefficients RC for each of the plurality of target models by referring to a ratio parameter (internal division ratio parameter RP) that defines the ratio of the plurality of target models and the target data TD. The calculation process of the sparse regression coefficients RC includes, for example, a preliminary regression coefficient calculation process S12-1 and a sparse regression coefficient calculation process S12-2.

[0079] (Step S12-1: Preliminary regression coefficient calculation process) First, in this step, the regression coefficient calculation unit 12 calculates preliminary regression coefficients for each of the plurality of target models by referring to the ratio parameters (internal ratio parameters RP) and the target data TD. As an example, the regression coefficient calculation unit 12 calculates the preliminary regression coefficients for the internal ratio parameters (internal ratio parameter vector) RP expressed as the following equation (13).

number

number

number

number

number

[0080] (Step S12-2: Sparse regression coefficient calculation process) Next, the regression coefficient calculation unit 12 calculates the sparse regression coefficients RC by a sparse linear regression process that refers to a pair of a predicted value obtained by applying the preliminary regression coefficients W to the target data and the target data. As an example, the regression coefficient calculation unit 12 calculates the preliminary regression coefficients W for the target model with index 1 among the plurality of target models. (1) By applying the above to the target data TD, the predicted value ~y (1) k and the predicted value ~y (1) k and the feature value ~x included in the target data k Paired with

number

number

[0081] (Step S13: Covariance calculation process) Subsequently, in step S13, the covariance calculation unit 13 calculates the target data TD expressed as the following equation (18):

number

number

number

number

number

number

[0082] (Step S14: Internal division ratio parameter calculation process) Subsequently, in step S14, the internal division ratio parameter calculation unit 14 calculates the target data TD expressed as the following equation (24):

number

number

number

number

number

number

number

[0083] (Step S17: Convergence determination process) Subsequently, in step S17, the convergence determination unit 17 determines whether or not the series of processes in the above-mentioned steps S12, S13, and S14 have converged. This may be expressed as determining whether or not the above-mentioned variational Bayes EM algorithm has converged, or as determining whether or not the calculation related to the interior ratio parameter RP in step S14 has converged. As an example, the convergence determination unit 17 determines whether or not the variational lower bound (VLB) J obtained by the following equation (31),

number

[0084] If the convergence determination unit 17 determines that the convergence has occurred, the process proceeds to the output process S15; if the convergence determination unit 17 determines that the convergence has not occurred, the process returns to the regression coefficient calculation process S12, and the calculation of the sparse regression coefficient RC is repeated.

[0085] (Step S15: Output process) If the convergence determination unit 17 determines that "convergence has occurred" in step S17, then in step S15 the output unit 15 outputs the sparse regression coefficients RC calculated by the regression coefficient calculation unit 12 in step S12 and the interior ratio parameters RP (learning result LR) calculated by the interior ratio parameter calculation unit 14 in step S14. In this way, if the convergence determination unit 17 determines that the calculation related to the interior ratio parameters RP has "converged," the output unit 15 outputs the learning result LR, thereby making it possible to output a suitable learning result LR.

[0086] In step S15, the output unit 15 outputs the hidden variable z calculated by the covariance calculation unit 13 in step S13. k The covariance parameter η of the prior distribution of the hidden variable z k The output unit 15 may be configured to further output the covariance matrix Ψ of the posterior distribution of the hidden variable z k The covariance parameter η of the prior distribution of and the hidden variable z k The covariance matrix Ψ of the posterior distribution of σ can be presented to the user.

[0087] In step S15, the output unit 15 may be configured to display graphs defined by the regression coefficients RC for at least two of the plurality of target models in a manner that allows them to be distinguished from one another.

[0088] 8 shows an example of a graph that the output unit 15 displays via the input / output unit 17A in this step. In the example shown in FIG. 8, the output unit 15 displays the sparse regression coefficients RC calculated for each of the multiple target models, which are expressed as the following equation (32), in the regression coefficient calculation process in step S12.

number

[0089] 8, the output unit 15 may refer to the correspondence relationship information CI and display what models the graph L1 and the graph L2 correspond to. In Fig. 8, the output unit 15 displays that the graph L1 is a model corresponding to the situation A, and the graph L2 is a model corresponding to the situation B.

[0090] (Example of processing executed by the prediction unit 18) The prediction unit 18 derives multiple prediction results PR by applying the regression coefficients RC for each of the multiple target models and the ratio parameters RP calculated by the ratio parameter calculation unit 14, which are determined according to the type of the inference data PD, to the inference data PD. As an example, when the inference data PD includes a feature amount (explanatory variable) x0, the prediction unit 18 derives the prediction result PR by applying the regression coefficients RC for each of the multiple target models and the ratio parameters RP determined according to the type of the inference data PD to the feature amount x0. In this case, an example of a graph including the prediction results RP displayed by the output unit 15 via the input / output unit 17A is shown in FIG. 9. FIG. 9 shows another example of a graph displayed by the output unit 15 via the input / output unit 17A.

[0091] In addition, the ratio parameter determined according to the type of inference data PD refers, as an example, to a ratio parameter learned (updated) by the above-mentioned learning process using learning data (target data) associated with a type that is the same as or similar to the type of the inference data PD, but this does not limit this exemplary embodiment.

[0092] For example, assume that each of the multiple target data TD is sales of drinking water at multiple stores. Also, a situation in which a campaign for the drinking water is being conducted is defined as situation A, and a situation in which the volume of the drinking water has been increased is defined as situation B. Examples of situations include, but are not limited to, a situation in which the appearance of the drinking water container has been changed, a situation in which a new store has opened near the store, a situation in which it is raining, and a situation in which all of these situations are included.

[0093] For example, when predicting sales of a drink at a certain store at a certain date and time x0, where a campaign for the drink is being carried out and the volume of the drink has been increased, the prediction unit 18 calculates the regression coefficient W of model 1 corresponding to situation A at the certain date and time x0. (1) The prediction result P1 is derived by applying the regression coefficient W of Model 2 corresponding to situation B at a certain date and time x0. (2) The prediction unit 18 derives the prediction result P2 by applying the ratio parameter μ (1) and the ratio parameter μ of Model 2 (2) The prediction result RP is derived by applying

[0094] In this way, the information processing device 1A according to this exemplary embodiment can generate a prediction result PR according to the type of inference data PD. Therefore, the information processing device 1A according to this exemplary embodiment can present a prediction result PR that has favorable explainability to the user.

[0095] 10, the output unit 15 may display a graph defined by the regression coefficients R of the respective models. FIG. 10 shows another example of a graph displayed by the output unit 15 via the input / output unit 17A. The output unit 15 displays a graph L1 defined by the regression coefficients R of model 1 corresponding to situation A, a graph L2 defined by the regression coefficients R of model 2 corresponding to situation B, and a graph L3 defined by the regression coefficients R of model 3 corresponding to situation C in a manner that allows them to be distinguished from one another. Situation C may include at least one of situations A and B, or may be a situation different from situations A and B. According to the information processing device 1A according to this exemplary embodiment, even with the above-described configuration, it is possible to present a prediction result PR having favorable explainability to the user.

[0096] (Effects of information processing device 1A) As described above, in the information processing device 1A, Obtain the target data and the number of models for multiple target models, calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio between the plurality of target models and the target data; Calculating the ratio parameter by referring to the target data and the regression coefficients; Outputting the regression coefficients and the ratio parameters The following configuration is adopted.

[0097] According to the above configuration, sparse regression coefficients for each of the plurality of target models are calculated by referring to the ratio parameters and the target data, so that it is possible to perform optimization processing (learning processing, update processing) for the ratio parameters while suppressing overlearning. Therefore, according to the above configuration, it is possible to suitably suppress overlearning in a configuration that uses scheduling parameters (ratio parameters).

[0098] Third Exemplary Embodiment A third exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same functions as those described in the above exemplary embodiment will be assigned the same reference numerals, and their description will be omitted as appropriate. The scope of application of each technical means employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. Furthermore, each technical means shown in each drawing referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise.

[0099] (Configuration of information processing device 2A) The configuration of the information processing device 2A will be described with reference to Fig. 11. Fig. 11 is a block diagram showing the configuration of the information processing device 2A. As shown in Fig. 11, the information processing device 2A includes a control unit 20A, a storage unit 25A, a communication unit 26A, and an input / output unit 27A.

[0100] The storage unit 25A stores data referenced by the control unit 20A, similar to the storage unit 15A described above. Examples of data stored in the storage unit 25A include, but are not limited to, inference data PD, learning results LR, and prediction results PR, as shown in Fig. 11. The inference data PD, correspondence information CI, learning results LR, and prediction results PR are as described above.

[0101] The communication unit 26A is an interface that transmits and receives data via a network, similar to the above-described communication unit 16A.

[0102] The input / output unit 27A is an interface that receives input of data and outputs data, similar to the above-described input / output unit 17A.

[0103] (Control unit 20A) The control unit 20A controls each of the components included in the information processing device 2 A. The control unit 20A also includes an acquisition unit 21, a prediction unit 22, and an output unit 23, as shown in FIG.

[0104] (Acquisition part 21) The acquisition unit 21 acquires the inference data PD and information on the type of the inference data PD. The acquisition unit 21 supplies the acquired inference data to the prediction unit 22.

[0105] (Prediction Section 22) The prediction unit 22 derives a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models and is determined according to the type of the inference data to the inference data PD acquired by the acquisition unit 21. Here, the regression coefficient and the ratio parameter are A regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The regression coefficients and ratio parameters are regression coefficients and ratio parameters learned by a learning process including the above. Furthermore, the ratio parameters determined according to the type of inference data refer, for example, to ratio parameters learned (updated) by a learning process using learning data (target data) associated with a type that is the same as or similar to the type of the inference data, but this does not limit the present exemplary embodiment. For example, the regression coefficients and ratio parameters are regression coefficients calculated by the information processing device 1A according to the present exemplary embodiment. The prediction unit 22 supplies the prediction result to the output unit 23.

[0106] (output unit 23) The output unit 23 outputs the prediction result PR by the prediction unit 22. Output by the output unit 23. Examples of the prediction result PR output by the output unit 23 are as described with reference to FIGS. 8 to 10. That is, the output unit 23 displays graphs showing at least two prediction results PR out of the multiple prediction results PR, the graphs being defined by regression coefficients, in a manner that allows the graphs to be distinguished from one another. Therefore, the output unit 23 can generate a prediction result PR having a range for the inference data PD (as an example, a prediction result having a range defined by the prediction results P1 and P2 shown in FIG. 9).

[0107] (Effects of information processing device 2) As described above, in the information processing device 2, Acquire inference data and information regarding the type of the inference data, deriving a prediction result by applying a regression coefficient for each of a plurality of target models and a ratio parameter that defines a ratio of the plurality of target models to the inference data, the ratio parameter being determined according to the type of the inference data; Outputting the prediction result by the prediction means In this configuration, The regression coefficients and the ratio parameters are regression coefficients and ratio parameters learned by a learning process including a regression coefficient calculation process that calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and learning data, and a ratio parameter calculation process that calculates the ratio parameters by referring to the learning data and the regression coefficients. The following configuration is adopted.

[0108] According to the above configuration, sparse regression coefficients for each of the plurality of target models are calculated by referring to the ratio parameters and the target data, and therefore, optimization processing (learning processing, update processing) for the ratio parameters is performed while suppressing overfitting. Therefore, according to the above configuration, overfitting is preferably suppressed in a configuration using scheduling parameters (ratio parameters). Furthermore, prediction can be preferably performed using regression coefficients in which overfitting is suppressed.

[0109] Fourth Exemplary Embodiment A fourth exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same functions as those described in the above exemplary embodiment will be assigned the same reference numerals, and their description will be omitted as appropriate. The scope of application of each technical means employed in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technical means employed in this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise. Furthermore, each technical means shown in each drawing referenced to explain this exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure, to the extent that no particular technical obstacles arise.

[0110] (Configuration of information processing device 1B) The configuration of the information processing device 1B will be described with reference to Fig. 12. Fig. 12 is a block diagram showing the configuration of the information processing device 1B. As shown in Fig. 12, the information processing device 1B includes a control unit 10B, a storage unit 15B, a communication unit 16A, and an input / output unit 17A. The communication unit 16A and the input / output unit 17A have been described above, and therefore description thereof will be omitted.

[0111] Similar to the above-described storage unit 15A, the storage unit 15B stores data referenced by the control unit 10B. Examples of data stored in the storage unit 15B include, but are not limited to, target data TD, ratio parameters RP, regression coefficients RC, correspondence information CI, condition information JI, constraints CC, learning results LR, inference data PD, and prediction results PR, as shown in Fig. 12. The target data TD, ratio parameters RP, regression coefficients RC, correspondence information CI, learning results LR, inference data PD, and prediction results PR are as described above, and therefore will not be described here.

[0112] (control unit 10B) The control unit 10B controls each component included in the information processing device 1B, similar to the control unit 10A described above. As shown in FIG. 12 , the control unit 10B includes an acquisition unit 11, a regression coefficient calculation unit 12, a ratio parameter calculation unit 14, an output unit 15, an initial value determination unit 16, a convergence determination unit 17, and a prediction unit 18. In this exemplary embodiment, the acquisition unit 11, the regression coefficient calculation unit 12, the ratio parameter calculation unit 14, the initial value determination unit 16, the convergence determination unit 17, and the prediction unit 18 respectively implement an acquisition unit, a regression coefficient calculation unit, a ratio parameter calculation unit, an initial value determination unit, a convergence determination unit, and a prediction unit. The output unit 15 implements a storage unit and an output unit. The acquisition unit 11, the output unit 15, the initial value determination unit 16, the convergence determination unit 17, and the prediction unit 18 are as described above, and therefore will not be described further.

[0113] The regression coefficient calculation unit 12 calculates a regression coefficient RC for each of the plurality of target models by referring to the ratio parameter RP that defines the ratio of the plurality of target models and the target data TD. In this exemplary embodiment, the regression coefficient calculation unit 12 calculates the regression coefficient RC for each of the plurality of target models by referring to the hidden variable z k The covariance parameter η of the prior distribution of and the hidden variable z k The regression coefficients RC are calculated using the least squares method without updating the covariance matrices Φ and Φ of the prior distributions.

[0114] The ratio parameter calculation unit 14 calculates the ratio parameter RP by referring to the target data TD and the regression coefficient RC. In this exemplary embodiment, the ratio parameter RP is not limited to an internal ratio parameter, and may be an external ratio parameter.

[0115] (Example of processing flow in information processing device 1B) 13 is a diagram showing an example of the flow of processing in the information processing device 1B according to this exemplary embodiment. The acquisition processing S11, the initial value determination processing S16, and the output processing S15 are as described above, and therefore descriptions thereof will be omitted.

[0116] (Step S12: Regression coefficient calculation process) Subsequently, in step S12, the regression coefficient calculation unit 12 calculates sparse regression coefficients RC for each of the plurality of target models by referring to a ratio parameter (internal division ratio parameter RP) that defines the ratio of the plurality of target models and the target data TD. The calculation process of the sparse regression coefficients RC includes, for example, a preliminary regression coefficient calculation process S12-1 and a sparse regression coefficient calculation process S12-2.

[0117] (Step S12-1: Preliminary regression coefficient calculation process) First, in this step, the regression coefficient calculation unit 12 calculates preliminary regression coefficients for each of the plurality of target models by referring to the ratio parameters (internal ratio parameters RP) and the target data TD. As an example, the regression coefficient calculation unit 12 calculates preliminary regression coefficients for the internal ratio parameters (internal ratio parameter vector) RP expressed as the following formula (A-13):

number

number

number

number

[0118] (Step S12-2: Sparse regression coefficient calculation process) Next, the regression coefficient calculation unit 12 calculates the sparse regression coefficients RC by a sparse linear regression process that refers to a pair of a predicted value obtained by applying the preliminary regression coefficients W to the target data and the target data. As an example, the regression coefficient calculation unit 12 calculates the preliminary regression coefficients W for the target model with index 1 among the plurality of target models. (1) By applying the above to the target data TD, the predicted value ~y (1) k and the predicted value ~y (1) k and the feature value ~x included in the target data k Paired with

number

number

[0119] (Step S14: Ratio parameter calculation process) Subsequently, in step S14, the ratio parameter calculation unit 14 calculates the target data TD expressed as the following equation (37):

number

number

number

number

number

number

[0120] As another example, the convergence determination unit 17 determines whether the value of the ratio parameter RP has converged. For example, in the nth convergence determination process of repeating a series of processes including the above-mentioned steps S12, S13, and S14, the convergence determination unit 17 compares the value of the ratio parameter RP at the (n-1)th time with the value of the ratio parameter RP at the nth time, and determines that the series of processes at the above-mentioned steps S12, S13, and S14 has converged if the absolute value of the difference between them is equal to or smaller than a predetermined threshold.

[0121] (Effects of information processing device 1B) As described above, in the information processing device 1B, the sparse regression coefficients R C are calculated using the least squares method. With this configuration, in the information processing device 1B as well, overlearning is preferably suppressed in a configuration using scheduling parameters (ratio parameters) similar to the information processing device 1A.

[0122] [Software implementation example] Some or all of the functions of the information processing devices 1, 1A, 2, 2A, and 1B (hereinafter also referred to as "each of the above devices") may be realized by hardware such as an integrated circuit (IC chip), or by software.

[0123] In the latter case, each of the above devices is realized by, for example, a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter referred to as computer C) is shown in Figure 14. Figure 14 is a block diagram showing the hardware configuration of computer C that functions as each of the above devices.

[0124] The computer C includes at least one processor C1 and at least one memory C2. The memory C2 stores a program P for causing the computer C to operate as each of the above-mentioned devices. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing the functions of each of the above-mentioned devices.

[0125] The processor C1 may be, for example, a central processing unit (CPU), a graphic processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a tensor processing unit (TPU), a quantum processor, a microcontroller, or a combination thereof. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination thereof.

[0126] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may also include a communication interface for transmitting and receiving data to and from other devices. The computer C may also include an input / output interface for connecting input / output devices such as a keyboard, mouse, display, and printer.

[0127] Furthermore, the program P can be recorded on a non-transitory tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can acquire the program P via such a recording medium M. The program P can also be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also acquire the program P via such a transmission medium.

[0128] [Appendix A] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0129] (Appendix A1) An acquisition means for acquiring target data and the number of models of a plurality of target models; a regression coefficient calculation means for calculating sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models; a ratio parameter calculation means for calculating the ratio parameters by referring to the target data and the regression coefficients; an output means for outputting the regression coefficients calculated by the regression coefficient calculation means and the ratio parameters calculated by the ratio parameter calculation means; An information processing device comprising:

[0130] (Appendix A2) The regression coefficient calculation means calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; Calculating the sparse regression coefficients by a sparse linear regression process that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. 10. The information processing device according to claim 1,

[0131] (Appendix A3) The obtaining means further obtains information regarding a prior distribution of the latent variables; The information processing device includes: Further comprising a covariance calculation means for calculating a covariance parameter of the prior distribution of the latent variables and a covariance matrix of the posterior distribution of the latent variables by referring to the target data, the ratio parameter, the regression coefficient, and information on the prior distribution of the latent variables; The ratio parameter calculation means calculates the ratio parameter by further referring to a covariance matrix of the posterior distribution of the hidden variables. 10. The information processing device according to claim 9, wherein the information processing device is a

[0132] (Appendix A4) The ratio parameter calculation means calculates the ratio parameters under constraints on the sum of the ratio parameters. 10. The information processing device according to claim 9, wherein the information processing device is a

[0133] (Appendix A5) The method further includes an initial value determining means for determining an initial value of the ratio parameter to be referenced by the regression coefficient calculating means. 1. An information processing device according to claim A4.

[0134] (Appendix A6) further comprising a convergence determination means for determining whether or not the calculation regarding the ratio parameter has converged; The output means outputs the regression coefficients and the ratio parameters when the convergence determination means determines that convergence has occurred. An information processing device according to any one of appendices A1 to A5.

[0135] (Appendix A7) The output means displays graphs defined by the regression coefficients for at least two of the plurality of target models in a manner that allows them to be distinguished from one another. An information processing device according to any one of appendices A1 to A6.

[0136] (Appendix A8) the acquisition means further acquires data for inference; The information processing device includes: a prediction means for deriving a plurality of prediction results by applying the regression coefficients for each of the plurality of target models to the inference data; It also has An information processing device according to any one of appendices A1 to A7.

[0137] (Appendix A9) an acquisition means for acquiring inference data and information regarding the type of the inference data; The inference data includes: regression coefficients for each of a plurality of target models; a ratio parameter that defines a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data; a prediction means for deriving a prediction result by applying an output means for outputting a prediction result by the prediction means; Equipped with The regression coefficients and the ratio parameters are a regression coefficient calculation process for calculating sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The information processing device is a regression coefficient and a ratio parameter learned by a learning process including the steps of:

[0138] (Appendix A10) The output means displays graphs showing at least two of the plurality of prediction results, the graphs being defined by the regression coefficients, in a manner that allows them to be distinguished from one another. 10. The information processing device according to claim 9.

[0139] [Appendix B] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0140] (Appendix B1) an acquisition process in which at least one processor acquires target data and the number of models of a plurality of target models; a regression coefficient calculation process in which the at least one processor calculates sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models; a ratio parameter calculation process in which the at least one processor calculates the ratio parameter by referring to the target data and the regression coefficient; an output process in which the at least one processor outputs the regression coefficients calculated by the regression coefficient calculation process and the ratio parameters calculated by the ratio parameter calculation process; An information processing method comprising:

[0141] (Appendix B2) In the regression coefficient calculation process, the at least one processor: calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; Calculating the sparse regression coefficients by a sparse linear regression process that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. 1. The information processing method described in Appendix B1.

[0142] (Appendix B3) In the obtaining process, the at least one processor further obtains information regarding a prior distribution of a hidden variable; The information processing device includes: The at least one processor further includes a covariance calculation process for calculating covariance parameters of the prior distribution of the latent variables and a covariance matrix of the posterior distribution of the latent variables by referring to the target data, the ratio parameters, the regression coefficients, and information on the prior distribution of the latent variables; In the ratio parameter calculation process, the at least one processor further refers to a covariance matrix of the posterior distribution of the hidden variables to calculate the ratio parameter. 1. The information processing method described in Appendix B2.

[0143] (Appendix B4) In the ratio parameter calculation process, the at least one processor calculates the ratio parameter under a constraint on the sum of the ratio parameters. The information processing method described in Appendix B3.

[0144] (Appendix B5) The at least one processor further executes an initial value determination process for determining an initial value of the ratio parameter referred to in the regression coefficient calculation process. The information processing method described in Appendix B4.

[0145] (Appendix B6) the at least one processor further performs a convergence determination process to determine whether the calculation regarding the ratio parameter has converged; In the output process, the at least one processor outputs the regression coefficients and the ratio parameters when it is determined that the convergence determination process has converged. 1. An information processing method according to any one of Appendices B1 to B5.

[0146] (Appendix B7) In the output process, the at least one processor displays graphs defined by the regression coefficients for at least two of the plurality of target models in a manner that allows them to be distinguished from one another. 10. An information processing method according to any one of appendices B1 to B6.

[0147] (Appendix B8) The acquisition process further acquires data for inference, The at least one processor performs a prediction process to derive a plurality of prediction results by applying the regression coefficients for each of the plurality of target models to the inference data. Run the following again: 10. An information processing method according to any one of appendices B1 to B7.

[0148] (Appendix B9) an acquisition process in which the at least one processor acquires inference data and information regarding the type of the inference data; The inference data includes: regression coefficients for each of a plurality of target models; a ratio parameter that defines a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data; a prediction process in which the at least one processor derives a prediction result by applying an output process in which the at least one processor outputs a prediction result obtained by the prediction process; Including, The regression coefficients and the ratio parameters are a regression coefficient calculation process in which the at least one processor calculates sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the training data; a ratio parameter calculation process in which the at least one processor calculates the ratio parameter by referring to the training data and the regression coefficients; The information processing method is a regression coefficient and a ratio parameter learned by a learning process including:

[0149] (Appendix B10) In the output process, the at least one processor displays graphs showing at least two of the plurality of prediction results, the graphs being defined by the regression coefficients, in a manner that allows them to be distinguished from one another. 1. The information processing method described in Appendix B9.

[0150] [Appendix C] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0151] (Appendix C1) A program that causes a computer to function as an information processing device, the program comprising: An acquisition means for acquiring target data and the number of models of a plurality of target models; a regression coefficient calculation means for calculating sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models; a ratio parameter calculation means for calculating the ratio parameters by referring to the target data and the regression coefficients; an output means for outputting the regression coefficients calculated by the regression coefficient calculation means and the ratio parameters calculated by the ratio parameter calculation means; An information processing program that functions as a

[0152] (Appendix C2) The regression coefficient calculation means calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; Calculating the sparse regression coefficients by a sparse linear regression process that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. An information processing program as described in Appendix C1.

[0153] (Appendix C3) The obtaining means further obtains information regarding a prior distribution of the latent variables; The information processing device includes: The computer and further functioning as a covariance calculation means for calculating a covariance parameter of the prior distribution of the latent variables and a covariance matrix of the posterior distribution of the latent variables by referring to the target data, the ratio parameter, the regression coefficient, and information on the prior distribution of the latent variables; The ratio parameter calculation means calculates the ratio parameter by further referring to a covariance matrix of the posterior distribution of the hidden variables. An information processing program as described in Appendix C2.

[0154] (Appendix C4) The ratio parameter calculation means calculates the ratio parameters under constraints on the sum of the ratio parameters. An information processing program as described in Appendix C3.

[0155] (Appendix C5) The computer The regression coefficient calculation means further functions as an initial value determination means for determining an initial value of the ratio parameter to be referenced by the regression coefficient calculation means. An information processing program as described in Appendix C4.

[0156] (Appendix C6) The computer further functioning as a convergence determination means for determining whether or not the calculation regarding the ratio parameter has converged; The output means outputs the regression coefficients and the ratio parameters when the convergence determination means determines that convergence has occurred. An information processing program according to any one of appendices C1 to C5.

[0157] (Appendix C7) The output means displays graphs defined by the regression coefficients for at least two of the plurality of target models in a manner that allows them to be distinguished from one another. An information processing program according to any one of appendices C1 to C6.

[0158] (Appendix C8) the acquisition means further acquires data for inference; The information processing device includes: The computer a prediction means for deriving a plurality of prediction results by applying the regression coefficients for each of the plurality of target models to the inference data; Further function as An information processing program according to any one of appendices C1 to C7.

[0159] (Appendix C9) The computer an acquisition means for acquiring inference data and information regarding the type of the inference data; The inference data includes: regression coefficients for each of a plurality of target models; a ratio parameter that defines a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data; a prediction means for deriving a prediction result by applying an output process for outputting the prediction result by the prediction means; It functions as The regression coefficients and the ratio parameters are a regression coefficient calculation process for calculating sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The information processing program is a regression coefficient and a ratio parameter learned by a learning process including:

[0160] (Appendix C10) The output means displays graphs showing at least two of the plurality of prediction results, the graphs being defined by the regression coefficients, in a manner that allows them to be distinguished from one another. An information processing program as described in Appendix C9.

[0161] [Appendix D] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0162] (Appendix D1) at least one processor, An acquisition process for acquiring target data and the number of models of multiple target models; a regression coefficient calculation process for calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; a ratio parameter calculation process for calculating the ratio parameters by referring to the target data and the regression coefficients; an output process for outputting the regression coefficients calculated by the regression coefficient calculation process and the ratio parameters calculated by the ratio parameter calculation process; An information processing device that executes the above. The information processing device may further include a memory, and the memory may store a program for causing the at least one processor to execute each of the processes.

[0163] (Appendix D2) In the regression coefficient calculation process, the at least one processor: calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; Calculating the sparse regression coefficients by a sparse linear regression process that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. 10. The information processing device according to claim 9, wherein the information processing device is a device for processing information.

[0164] (Appendix D3) In the obtaining process, the at least one processor further obtains information regarding a prior distribution of a hidden variable; The at least one processor: Further performing a covariance calculation process to calculate a covariance parameter of the prior distribution of the latent variable and a covariance matrix of the posterior distribution of the latent variable by referring to the target data, the ratio parameter, the regression coefficient, and information on the prior distribution of the latent variable; In the ratio parameter calculation process, the at least one processor further refers to a covariance matrix of the posterior distribution of the hidden variables to calculate the ratio parameter. 10. The information processing device according to claim 9, wherein the information processing device is an information processing device according to claim 1, wherein

[0165] (Appendix D4) In the ratio parameter calculation process, the at least one processor calculates the ratio parameter under a constraint on the sum of the ratio parameters. 10. The information processing device according to claim 9, wherein the information processing device is an information processing device according to claim 1, wherein

[0166] (Appendix D5) the at least one processor: An initial value determination process is further performed to determine the initial value of the ratio parameter referred to in the regression coefficient calculation process. 10. The information processing device according to claim 9, wherein the information processing device is an information processing device according to claim 1, wherein

[0167] (Appendix D6) the at least one processor: further performing a convergence determination process to determine whether or not the calculation regarding the ratio parameter has converged; In the output process, the at least one processor outputs the regression coefficients and the ratio parameters when it is determined that the convergence determination process has converged. An information processing device according to any one of appendices D1 to D5.

[0168] (Appendix D7) In the output process, the at least one processor displays graphs defined by the regression coefficients for at least two of the plurality of target models in a manner that allows them to be distinguished from one another. An information processing device according to any one of appendices D1 to D6.

[0169] (Appendix D8) The acquisition process further acquires data for inference, The information processing device includes: the at least one processor: a prediction process for deriving a plurality of prediction results by applying the regression coefficients for each of the plurality of target models to the inference data; Run the following again: An information processing device according to any one of appendices D1 to D7.

[0170] (Appendix D9) The at least one processor: an acquisition process for acquiring inference data and information regarding the type of the inference data; The inference data includes: regression coefficients for each of a plurality of target models; a ratio parameter that defines a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data; a prediction process for deriving a prediction result by applying an output process for outputting a prediction result obtained by the prediction process; Prepare, execute, The regression coefficients and the ratio parameters are a regression coefficient calculation process for calculating sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The information processing device is configured to process the regression coefficients and ratio parameters learned by a learning process that executes the above.

[0171] (Appendix D10) In the output process, the at least one processor displays graphs showing at least two of the plurality of prediction results, the graphs being defined by the regression coefficients, in a manner that allows them to be distinguished from one another. 10. The information processing device according to claim 9,

[0172] [Appendix E] This disclosure includes the techniques described in the following appendices. However, the present invention is not limited to the techniques described in the following appendices, and various modifications are possible within the scope of the claims.

[0173] (Appendix E1) A program that causes a computer to function as an information processing device, the program comprising: An acquisition process for acquiring target data and the number of models of multiple target models; a regression coefficient calculation process for calculating sparse regression coefficients for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; a ratio parameter calculation process for calculating the ratio parameters by referring to the target data and the regression coefficients; an output process for outputting the regression coefficients calculated by the regression coefficient calculation process and the ratio parameters calculated by the ratio parameter calculation process; A non-transitory recording medium on which an information processing program for executing the above is recorded. [Explanation of symbols]

[0174] 1, 1A, 2, 2A, 1B Information processing device 11, 21 Acquisition Department 12 Regression coefficient calculation section 13 Covariance calculation part 14 Internal division parameter calculation section 15, 23 Output section 16 Initial value determination section 17 Convergence judgment section 18, 22 Prediction Section TD Target Data RP ratio parameter RC regression coefficient DI distribution information LR learning results PD inference data PR prediction results

Claims

1. An acquisition means for acquiring target data and the number of models of a plurality of target models; a regression coefficient calculation means for calculating sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models; a ratio parameter calculation means for calculating the ratio parameters by referring to the target data and the regression coefficients; an output means for outputting the regression coefficients calculated by the regression coefficient calculation means and the ratio parameters calculated by the ratio parameter calculation means; An information processing device comprising:

2. The regression coefficient calculation means calculating a preliminary regression coefficient for each of the plurality of target models by referring to a ratio parameter that defines a ratio of the plurality of target models and the target data; Calculating the sparse regression coefficients by a sparse linear regression process that references a pair of a predicted value obtained by applying the preliminary regression coefficients to the target data and the target data. The information processing device according to claim 1 .

3. The obtaining means further obtains information regarding a prior distribution of the latent variables; The information processing device includes: Further comprising a covariance calculation means for calculating a covariance parameter of the prior distribution of the latent variables and a covariance matrix of the posterior distribution of the latent variables by referring to the target data, the ratio parameter, the regression coefficient, and information on the prior distribution of the latent variables; The ratio parameter calculation means calculates the ratio parameter by further referring to a covariance matrix of the posterior distribution of the hidden variables. The information processing device according to claim 2 .

4. The ratio parameter calculation means calculates the ratio parameters under constraints on the sum of the ratio parameters. The information processing device according to claim 3 .

5. further comprising a convergence determination means for determining whether or not the calculation regarding the ratio parameter has converged; The output means outputs the regression coefficients and the ratio parameters when the convergence determination means determines that convergence has occurred. The information processing device according to claim 1 .

6. The output means displays graphs defined by the regression coefficients for at least two of the plurality of target models in a manner that allows them to be distinguished from one another. The information processing device according to claim 1 .

7. the acquisition means further acquires data for inference; The information processing device includes: a prediction means for deriving a plurality of prediction results by applying the regression coefficients for each of the plurality of target models to the inference data; It also has The information processing device according to claim 1 .

8. an acquisition means for acquiring inference data and information regarding the type of the inference data; The inference data includes: regression coefficients for each of a plurality of target models; a ratio parameter that defines a ratio of the plurality of target models, the ratio parameter being determined according to the type of the inference data; a prediction means for deriving a prediction result by applying an output means for outputting a prediction result by the prediction means; Equipped with The regression coefficients and the ratio parameters are a regression coefficient calculation process for calculating sparse regression coefficients for each of a plurality of target models by referring to the ratio parameters and the learning data; and A ratio parameter calculation process for calculating the ratio parameters by referring to the learning data and the regression coefficients. The information processing device is configured to process the regression coefficients and ratio parameters learned by a learning process including the steps of:

9. The computer Obtaining target data and a model number of a plurality of target models; calculating sparse regression coefficients for each of the plurality of target models by referring to the target data and a ratio parameter that defines a ratio of the plurality of target models; calculating the ratio parameter by referring to the target data and the regression coefficients; outputting the regression coefficients and the ratio parameters; An information processing method comprising:

10. 2. A program for causing a computer to function as the information processing device according to claim 1, the program causing a computer to function as the acquisition means, the regression coefficient calculation means, the ratio parameter calculation means, and the output means.

Images