Information processing device, relearning system, relearning method, and relearning program

Abstract

The information processing device (100) includes an acquisition unit (120) that acquires an estimation target image (10) and a trained model, an estimation unit (130) that performs estimation using the estimation target image (10) and the trained model, a generation unit (140) that generates a heat map (11) using the estimation result, an extraction unit (150) that extracts a plurality of features based on the heat map (11), and a re-learning processing unit (180). The generation unit (140) generates a plurality of changed images by performing a change for each feature using the estimation target image (10). The estimation unit (130) performs estimation using the changed images (21 to 26) and the trained model. The generation unit (140) generates heat maps (21 a to 26 a) using a plurality of estimation results. If the estimation basis is incorrect, the re-learning processing unit (180) uses the training data to generate re-learned data in which the feature indicated by the estimation basis is changed, and re-learns the trained model.

Description

[Technical field]

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

[0002] In the field of machine learning, HITL (Human-in-the-Loop) is known. HITL means that feedback from humans is included in the learning process. Technology related to HITL has been proposed (see Patent Document 1). The image recognition device in Patent Document 1 corrects the attention map based on a human correction operation. This allows the attention map to be corrected by human knowledge. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-22368 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there are cases where a trained model does not make inferences based on correct inference grounds. Using such a trained model is problematic.

[0005] The objective of the present disclosure is to enable a trained model to perform estimation based on correct estimation grounds. [Means for solving the problem]

[0006] An information processing device according to one aspect of the present disclosure is provided. The information processing device includes an acquisition unit that acquires an estimation target image and a trained model, an estimation unit that performs estimation using the estimation target image and the trained model, a generation unit that generates a heat map showing the grounds of the estimation result using the estimation result, an extraction unit that extracts a plurality of features based on an area in the estimation target image that corresponds to a location of the grounds of the estimation result shown in the heat map and is an area in the estimation target image, and a re-learning processing unit. When the plurality of features are extracted, the generation unit generates a plurality of changed images by making changes for each of the features using the estimation target image. When the plurality of changed images are generated, the estimation unit performs estimation using the plurality of changed images and the trained model. When estimation is performed using the plurality of changed images and the trained model, the generation unit generates a plurality of heat maps using a plurality of estimation results. The acquisition unit acquires learning data when the estimation grounds are incorrect based on the plurality of changed images and the plurality of heat maps, or based on the estimation target image, the heat map, the plurality of changed images, and the plurality of heat maps. The re-learning processing unit uses the training data to generate re-learned data in which the features indicated by the estimation grounds have been changed, and re-learns the trained model using the re-learned data. Effect of the Invention

[0007] According to the present disclosure, a trained model can perform estimation based on correct estimation grounds. [Brief description of the drawings]

[0008] [Figure 1] FIG. 2 is a diagram illustrating hardware included in an information processing device according to the first embodiment. [Diagram 2] 1 is a block diagram showing functions of an information processing device according to a first embodiment; [Diagram 3] 5A to 5C are diagrams showing specific examples of an estimation target image and a heat map according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of a generation process according to the first embodiment. [Diagram 5] FIG. 4 is a diagram showing a specific example of output information according to the first embodiment. [Figure 6] FIG. 11 is a diagram showing a specific example (part 1) of the process of generating estimation basis information according to the first embodiment. [Figure 7] FIG. 11 is a diagram showing a specific example (part 2) of the process of generating estimation basis information according to the first embodiment. [Figure 8] 4 is a flowchart showing an example (part 1) of processing executed by the information processing device according to the first embodiment. [Figure 9] 13 is a flowchart showing a second example of the process executed by the information processing device according to the first embodiment. [Figure 10] FIG. 11 is a block diagram showing the functions of an information processing device according to a second embodiment. [Figure 11] FIG. 13 is a diagram showing a specific example of a generation process according to the second embodiment. [Figure 12] 13 is a flowchart showing an example (part 1) of processing executed by the information processing device according to the second embodiment. [Figure 13] 13 is a flowchart showing a second example of the process executed by the information processing device according to the second embodiment. [Figure 14] FIG. 13 is a diagram illustrating a re-learning system according to a third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an embodiment will be described with reference to the drawings. The following embodiment is merely an example, and various modifications are possible within the scope of the present disclosure.

[0010] Embodiment 1 1 is a diagram showing hardware included in an information processing device according to the first embodiment. The information processing device 100 is a computer. The information processing device 100 is a device that executes a relearning method. The information processing device 100 includes a processor 101, a volatile storage device 102, and a non-volatile storage device 103.

[0011] The processor 101 controls the entire information processing device 100. For example, the processor 101 is a central processing unit (CPU) or a field programmable gate array (FPGA). The processor 101 may be a multiprocessor. The information processing device 100 may also include a processing circuit.

[0012] The volatile storage device 102 is a main storage device of the information processing device 100. For example, the volatile storage device 102 is a random access memory (RAM). The non-volatile storage device 103 is an auxiliary storage device of the information processing device 100. For example, the non-volatile storage device 103 is a hard disk drive (HDD) or a solid state drive (SSD).

[0013] Next, functions of the information processing device 100 will be described. 2 is a block diagram showing functions of the information processing device of embodiment 1. The information processing device 100 includes a storage unit 110, an acquisition unit 120, an estimation unit 130, a generation unit 140, an extraction unit 150, an output unit 160, an identification unit 170, and a relearning processing unit 180.

[0014] The storage unit 110 may be realized as a storage area secured in the volatile storage device 102 or the non-volatile storage device 103 . A part or all of the acquisition unit 120, the estimation unit 130, the generation unit 140, the extraction unit 150, the output unit 160, the identification unit 170, and the relearning processing unit 180 may be realized by a processing circuit. Also, a part or all of the acquisition unit 120, the estimation unit 130, the generation unit 140, the extraction unit 150, the output unit 160, the identification unit 170, and the relearning processing unit 180 may be realized as a module of a program executed by the processor 101. For example, the program executed by the processor 101 is also referred to as a relearning program. For example, the relearning program is recorded on a recording medium.

[0015] The storage unit 110 stores various information. The acquisition unit 120 acquires the estimation target image. For example, the acquisition unit 120 acquires the estimation target image from the storage unit 110. Also, for example, the acquisition unit 120 acquires the estimation target image from an external device. For example, the external device is a cloud server. Note that the illustration of the external device is omitted. Also, for example, the acquisition unit 120 acquires the estimation target image obtained by a user's input operation. Note that the estimation target image is an image of the estimation target. The acquiring unit 120 acquires the trained model. For example, the acquiring unit 120 acquires the trained model from the storage unit 110 or an external device.

[0016] The estimation unit 130 performs estimation using the estimation target image and the trained model. In detail, the estimation unit 130 inputs the estimation target image to the trained model, and the trained model outputs an estimation result.

[0017] The generating unit 140 uses the estimation result to generate a heat map showing the basis of the estimation result. In particular, the generating unit 140 generates the heat map using XAI (Explainable Artificial Intelligence) and the estimation result. For example, the generating unit 140 generates the heat map using Grad-CAM (Gradient weighted Class Activation Mapping). Here, a specific example of the estimation target image and the generated heat map is shown.

[0018] Fig. 3 is a diagram showing a specific example of an estimation target image and a heat map according to embodiment 1. Fig. 3 shows an estimation target image 10 and a heat map 11. Furthermore, a range 11a indicates a location on which the estimation result is based.

[0019] The extraction unit 150 extracts a plurality of features based on the region 10a corresponding to the location of the basis of the estimation result in the estimation target image 10. For example, the extraction unit 150 extracts a light orange color based on the region 10a. The extraction unit 150 extracts gray, which is the color of the surroundings in the region 10a. In other words, the extraction unit 150 extracts gray, which is the color of the parts other than the fingers (i.e., the background) in the region 10a. The extraction unit 150 extracts that the brightness of the hand part is 75% based on the region 10a. The extraction unit 150 extracts that the brightness of the surroundings in the region 10a is 50%. The extraction unit 150 extracts high-frequency components based on the region 10a. For example, the high-frequency components are fine lines such as wrinkles. The extraction unit 150 extracts low-frequency components based on the region 10a. For example, the low-frequency components are thick lines such as fingers and hands.

[0020] Also, for example, the feature is a mid-frequency component. The extraction unit 150 can extract high-frequency components, mid-frequency components, and low-frequency components by using multi-resolution analysis. In this way, the extraction unit 150 may extract frequency components as features based on the region 10a.

[0021] Furthermore, for example, the feature is a line, a shape, a specific shape, etc. For example, the shape is a circle. For example, the specific shape is a finger shape, a hand shape, etc. The extraction unit 150 can extract the line, shape, specific shape, etc. by using a trained model, image analysis technology, etc.

[0022] When multiple features are extracted, the generating unit 140 generates multiple modified images by modifying each feature using the estimation target image. Note that a modified image is an image in which the feature has been modified or erased. The generation process will be described with reference to the drawings.

[0023] FIG. 4 is a diagram illustrating an example of the generation process according to the first embodiment. The generation unit 140 generates an image of a color opposite to the characteristic color in terms of hue, using the estimation target image 10. Specifically, the generation unit 140 generates a modified image 21 in which light orange (i.e., the color of the hand) is changed to light blue. In this way, the modified image 21 including the light blue hand is generated. Note that the information processing device 100 can eliminate the influence of the characteristic by generating an image of the opposite color.

[0024] The generating unit 140 generates an image having the same color as the surrounding color in the region 10a, using the estimation target image 10. Specifically, the generating unit 140 generates a modified image 22 having the same color as the surrounding color in the region 10a (i.e., gray). In this way, a modified image 22 including a gray hand is generated. Note that, as described later, the information processing device 100 can check whether the surrounding color in the region 10a is the estimation basis by generating the modified image 22.

[0025] The generation unit 140 generates an image with a brightness opposite to the brightness of the characteristic, using the estimation target image 10. Specifically, the generation unit 140 generates a modified image 23 with a brightness (25%=100-75) opposite to the brightness of the characteristic (i.e., 75%). This generates a modified image 23 with a brightness of 25% for the hand. Note that the information processing device 100 can eliminate the influence of the characteristic by generating an image with the opposite brightness.

[0026] The generating unit 140 generates an image with the same brightness as the brightness of the surroundings in the region 10a, using the estimation target image 10. Specifically, the generating unit 140 generates a modified image 24 with the same brightness as the brightness of the surroundings in the region 10a (i.e., 50%). In this way, a modified image 24 with a brightness of 50% for the hand is generated. Note that, as described later, by generating the modified image 24, the information processing device 100 can confirm whether the brightness of the surroundings in the region 10a is the basis for estimation.

[0027] The generating unit 140 uses the estimation target image 10 to generate a modified image 25 from which characteristic high frequency components have been removed. The generating unit 140 uses the estimation target image 10 to generate a modified image 26 from which characteristic low-frequency components have been removed.

[0028] Here, the term "removal" includes the meaning of weakening frequency components. As described later, the modified image is input to a trained model. If the trained model does not read frequency components, the frequency components do not need to be removed. Therefore, the generating unit 140 may generate a modified image in which the frequency components are weakened.

[0029] The estimation unit 130 performs estimation using a plurality of modified images and the trained model. In detail, the estimation unit 130 inputs each of the plurality of modified images to the trained model, and the trained model outputs a plurality of estimation results.

[0030] The generating unit 140 generates a plurality of heat maps using a plurality of estimation results. In particular, the generating unit 140 generates the plurality of heat maps using a plurality of estimation results and the XAI.

[0031] The output unit 160 outputs the estimation target image 10, the heat map 11, the multiple modified images, and the multiple heat maps. For example, the output unit 160 outputs the estimation target image 10 and the like to a display of the information processing device 100. Also, for example, the output unit 160 outputs the estimation target image 10 and the like to an external device. Here, examples of an estimation target image 10, a heat map 11, a number of modified images, and the heat maps are shown.

[0032] Fig. 5 is a diagram showing a specific example of output information according to the embodiment 1. Fig. 5 shows an estimation target image 10, a heat map 11, modified images 21 to 26, and heat maps 21a to 26a.

[0033] For example, by looking at the output information displayed on the display, the user can analyze that the high frequency components (i.e., wrinkles) are the basis for the estimation. That is, among heat maps 21a-26a, only heat map 25a does not show the location of the basis for the estimation result. In modified image 25, the high frequency components have been removed. Therefore, heat map 25a does not show the location of the basis for the estimation result. In other words, since modified images 21-24, 26 contain high frequency components, heat maps 21a-24a, 26a show the location of the basis for the estimation result. Therefore, the user can analyze that the high frequency components are the basis for the estimation.

[0034] Furthermore, the output unit 160 may output the estimation basis information. The estimation basis information is generated based on the estimation target image, a heat map corresponding to the estimation target image, a plurality of modified images, and a plurality of heat maps. The output of the estimation basis information will be described below.

[0035] 6 is a diagram showing a specific example (part 1) of the process of generating estimation basis information according to the embodiment 1. FIG. 6 shows a modified image 30 and a heat map 30a corresponding to the modified image 30. The identification unit 170 identifies, from among a plurality of heat maps corresponding to a plurality of modified images, the heat map that is most different from the heat map 11 corresponding to the estimation target image 10. In this way, the heat map 30a is identified.

[0036] The generating unit 140 generates an image 40 indicating the estimation basis based on the difference between the estimation target image 10 and the modified image 30 corresponding to the heat map 30a. The image 40 is also referred to as estimation basis information. The output unit 160 outputs the estimation target image, a heat map corresponding to the estimation target image, a plurality of modified images, a plurality of heat maps corresponding to the plurality of modified images, and the image 40. Alternatively, the output unit 160 may output only the image 40.

[0037] The information processing device 100 outputs the image 40. Therefore, by visually checking the image 40, the user can easily identify the basis for the estimation.

[0038] Furthermore, the information processing device 100 may output the estimation basis information by the following process, which will be described with reference to the drawings.

[0039] Fig. 7 is a diagram showing a specific example (part 2) of the generation process of estimation basis information according to the first embodiment. Fig. 7 shows a modified image 51 and a heat map 51a corresponding to the modified image 51. Fig. 7 also shows a modified image 52 and a heat map 52a corresponding to the modified image 52. The identification unit 170 identifies, from among a plurality of heat maps corresponding to a plurality of modified images, a plurality of heat maps that are different from the heat map 11 corresponding to the estimation target image 10. In this way, the heat maps 51a and 52a are identified. The plurality of heat maps are also referred to as a plurality of specified heat maps.

[0040] The generating unit 140 generates a plurality of images based on the difference between the estimation target image 10 and a plurality of modified images corresponding to the identified plurality of heat maps. For example, the generating unit 140 generates an image 61 indicating the estimation basis based on the difference between the estimation target image 10 and the modified image 51 corresponding to the heat map 51a. Also, for example, the generating unit 140 generates an image 62 indicating the estimation basis based on the difference between the estimation target image 10 and the modified image 52 corresponding to the heat map 52a. The images 61 and 62 are also referred to as estimation basis information.

[0041] Furthermore, the generating unit 140 generates a plurality of images based on the difference between the heat map 11 corresponding to the estimation target image 10 and the specified plurality of heat maps. For example, the generating unit 140 generates an image 61a based on the difference between the heat map 11 and the heat map 51a. For example, the generating unit 140 generates an image 62a based on the difference between the heat map 11 and the heat map 52a. The plurality of images are also referred to as a plurality of estimation basis degree images.

[0042] The output unit 160 outputs the estimation basis information. For example, the output unit 160 outputs images 61 and 62. By visually checking the images 61 and 62, the user can recognize that there are multiple estimation grounds.

[0043] Furthermore, the output unit 160 outputs a plurality of estimation grounds degree images. For example, the output unit 160 outputs images 61a and 62a. The user can recognize the degree of estimation grounds by visually checking the images 61a and 62a. For example, an area 61a1 in the image 61a is an area indicating the estimation grounds. An area 62a1 in the image 62a is an area indicating the estimation grounds. The color of the area 62a1 is darker than the color of the area 61a1. Therefore, the user can recognize that the estimation grounds indicated by the image 62 have a stronger degree of estimation grounds than the estimation grounds indicated by the image 61.

[0044] Here, the user can recognize the basis for the estimation by visually checking the output information of the information processing device 100. For example, as shown in FIG. 5, the user can recognize the basis for the estimation by looking at the output information displayed on the display. Also, for example, as shown in FIGS. 6 and 7, the user can recognize the basis for the estimation by looking at the output information displayed on the display. In this way, the user can recognize the basis for the estimation by visually checking the output information. Therefore, the user can determine whether the basis for the estimation is incorrect or not.

[0045] When the user determines that the estimation basis is incorrect, the user inputs an instruction to change the feature indicated by the estimation basis in the learning data and an instruction to re-learn to the information processing device 100. As a result, the acquisition unit 120 acquires the change instruction and the re-learning instruction. The acquisition unit 120 acquires the learning data. For example, the acquisition unit 120 acquires the learning data from the storage unit 110 or an external device. Note that, for example, the learning data is data used when the trained model is generated. The re-learning processing unit 180 uses the learning data to generate re-learned data in which the feature indicated by the estimation basis has been changed. The re-learning processing unit 180 re-learns the trained model using the re-learned data. As a result, the location of the estimation basis according to the trained model is changed to the correct location.

[0046] The above process will be described using a specific example. It is assumed that the feature indicated by the estimation basis is a high-frequency component. The user determines that the estimation basis is not a high-frequency component. The user inputs an instruction to change the high-frequency component in the learning data and an instruction to re-learn to the information processing device 100. The re-learning processing unit 180 uses the learning data to generate re-learned data in which the high-frequency component has been changed. For example, the re-learning processing unit 180 generates re-learned data obtained by removing the high-frequency component in the learning data. Also, for example, the re-learning processing unit 180 generates re-learned data obtained by weakening the high-frequency component in the learning data. The re-learning processing unit 180 re-learns the learned model using the re-learned data. As a result, the learned model does not perform estimation based on the high-frequency component. Therefore, the location of the estimation basis by the learned model is changed to the correct location.

[0047] In this way, the acquisition unit 120 acquires learning data when the estimation basis is incorrect based on the multiple modified images and multiple heat maps, or the estimation basis information. The re-learning processing unit 180 uses the learning data to generate re-learning data. The re-learning processing unit 180 uses the re-learning data to re-learn the trained model.

[0048] Next, the process executed by the information processing device 100 will be described with reference to a flowchart. FIG. 8 is a flowchart illustrating an example (part 1) of the process executed by the information processing device according to the first embodiment. (Step S11) The acquisition unit 120 acquires an estimation target image and a trained model. (Step S12) The estimation unit 130 performs estimation using the estimation target image and the trained model. (Step S13) The generating unit 140 generates a heat map using the estimation result. (Step S14) The extraction unit 150 extracts a plurality of features based on the area corresponding to the location of the grounds of the estimation result shown in the heat map. Note that the area is an area in the estimation target image.

[0049] (Step S15) The generating unit 140 generates a plurality of modified images based on a plurality of features. (Step S16) The estimation unit 130 performs estimation using the multiple changed images and the trained model. (Step S17) The generating unit 140 generates a plurality of heat maps using a plurality of estimation results. (Step S18) The output unit 160 outputs the estimation target image, a heat map corresponding to the estimation target image, a plurality of modified images, and a plurality of heat maps corresponding to the plurality of modified images. Note that the output unit 160 does not have to output the estimation target image and the heat map corresponding to the estimation target image. Then, the process proceeds to step S21.

[0050] FIG. 9 is a flowchart illustrating a second example of the process executed by the information processing device according to the first embodiment. (Step S21) The acquiring unit 120 acquires an instruction to change a feature indicated by the estimation grounds and an instruction to re-learn in the learning data. (Step S22) The acquiring unit 120 acquires learning data. (Step S23) The re-learning processing unit 180 uses the learning data to generate re-learned data in which the feature indicated by the estimation basis has been changed. (Step S24) The re-learning processing unit 180 re-learns the trained model by using the re-learned data.

[0051] According to the first embodiment, when the estimation basis is incorrect, the information processing device 100 generates re-learned data in which the features indicated by the estimation basis have been changed. Then, the information processing device 100 re-learns the trained model using the re-learned data. As a result, the trained model no longer performs estimation based on the features indicated by the estimation basis. In this manner, the information processing device 100 re-learns the trained model so that the trained model performs estimation based on the correct estimation basis. Thus, the information processing device 100 can cause the trained model to perform estimation based on the correct estimation basis.

[0052] Embodiment 2 Next, a description will be given of embodiment 2. In embodiment 2, differences from embodiment 1 will be mainly described. In embodiment 2, descriptions of matters common to embodiment 1 will be omitted.

[0053] 10 is a block diagram showing functions of an information processing device according to embodiment 2. An information processing device 100a includes a generating unit 140a, an identifying unit 170a, and a relearning processing unit 180a. The functions of the generating unit 140a, the identifying unit 170a, and the relearning processing unit 180a will be described later. Next, the generation process performed by the information processing device 100a will be described with reference to the drawings. Fig. 11 is a diagram showing a specific example of the generation process according to the embodiment 2. Fig. 11 shows a modified image 70 and a heat map 70a corresponding to the modified image 70.

[0054] The identification unit 170a identifies, from among a plurality of heat maps corresponding to a plurality of modified images, the heat map that is most different from the heat map 11 corresponding to the estimation target image 10. In this way, the heat map 70a is identified.

[0055] The generating unit 140a generates the features changed when the modified image 70 corresponding to the heat map 70a is generated as character information 80. For example, the modified image 70 is an image from which high-frequency components, which are features, have been removed. Therefore, the generating unit 140a generates character information 80 indicating the high-frequency components. The character information 80 is also referred to as estimation basis information.

[0056] The output unit 160 may output the text information 80. This allows the user to intuitively recognize the basis for the estimation.

[0057] Next, the process relating to re-learning in the second embodiment will be described. The acquiring unit 120 acquires correct answer information indicating a correct estimation basis. For example, the acquiring unit 120 acquires the correct answer information from the storage unit 110 or an external device. The re-learning processing unit 180a determines whether the estimation basis is incorrect based on the estimation basis information and the correct answer information. If the estimation basis is incorrect, the acquisition unit 120 acquires learning data. The re-learning processing unit 180a uses the learning data to generate re-learned data in which the features indicated by the estimation basis information have been changed. For example, the re-learning processing unit 180a uses the learning data to generate re-learned data in which the features (e.g., high-frequency components) indicated by the estimation basis information (e.g., character information 80) have been changed. The re-learning processing unit 180a re-learns the learned model using the re-learned data.

[0058] Next, the process executed by the information processing device 100a will be described with reference to a flowchart. Fig. 12 is a flowchart showing an example (part 1) of the process executed by the information processing device of the second embodiment. The process in Fig. 12 differs from the process in Fig. 8 in that steps S18a and 18b are executed. Therefore, steps S18a and 18b will be described in Fig. 12. Descriptions of the process other than steps S18a and 18b will be omitted.

[0059] (Step S18a) The identification unit 170a identifies the heat map that is most different from the heat map 11 corresponding to the estimation target image 10 from among the heat maps corresponding to the modified images. (Step S18b) The generating unit 140a generates, as text information, the features that have been changed when the changed image corresponding to the identified heat map has been generated, thereby generating estimated grounds information.

[0060] Fig. 13 is a flowchart showing an example (part 2) of the process executed by the information processing device of the second embodiment. The process in Fig. 13 differs from the process in Fig. 9 in that steps S21a, 21b, and 23a are executed. Therefore, steps S21a, 21b, and 23a will be described in Fig. 13. Descriptions of the processes other than steps S21a, 21b, and 23a will be omitted.

[0061] (Step S21a) The acquiring unit 120 acquires correct answer information. (Step S21b) The re-learning processing unit 180a judges whether the estimation basis is incorrect based on the estimation basis information and the correct answer information. If the estimation basis is incorrect, the process proceeds to step S22. If the estimation basis is correct, the process ends. (Step S23a) The re-learning processing unit 180a uses the learning data to generate re-learned data in which the feature indicated by the estimation basis information has been changed.

[0062] According to the second embodiment, when the estimation basis is incorrect, the information processing device 100a generates re-learned data in which the characteristics indicated by the estimation basis are changed. Then, the information processing device 100a re-learns the trained model using the re-learned data. As a result, the trained model no longer performs estimation based on the characteristics indicated by the estimation basis. In this manner, the information processing device 100a re-learns the trained model so that the trained model performs estimation based on the correct estimation basis. Therefore, the information processing device 100a can cause the trained model to perform estimation based on the correct estimation basis.

[0063] In the above, a case where one estimation basis is specified has been described. When a plurality of estimation basis is specified, the information processing device 100a may generate a plurality of pieces of character information. In detail, the generating unit 140a generates a plurality of features that have been changed when a plurality of changed images corresponding to a plurality of heat maps are generated as estimation basis information that is a plurality of pieces of character information. The process will be specifically described with reference to FIG. 7. The generating unit 140a generates a plurality of features that have been changed when a plurality of changed images 51, 52 corresponding to the heat maps 51a, 52a are generated as a plurality of pieces of character information. The output unit 160 may output the plurality of pieces of character information as estimation basis information. This allows the user to intuitively recognize a plurality of estimation basis.

[0064] When the estimation basis information is generated, the re-learning processing unit 180a determines whether or not there is an error in the estimation basis based on the estimation basis information and the correct answer information. When there is an error in the estimation basis, the acquisition unit 120 acquires learning data. The re-learning processing unit 180a uses the learning data to generate re-learning data in which, among multiple features indicated by multiple pieces of character information, a feature having an error in the estimation basis is changed. The re-learning processing unit 180a uses the re-learning data to re-learn the learned model.

[0065] As a result, the information processing device 100a retrains the trained model using retrained data in which the feature with the erroneous estimation basis among the multiple features has been changed. Therefore, the retrained trained model will no longer perform estimation based on the erroneous estimation basis. Therefore, the information processing device 100a can cause the trained model to perform estimation based on the correct estimation basis.

[0066] Third embodiment In the first and second embodiments, the case where the invention is realized by one information processing device has been described. The first and second embodiments may be realized by a re-learning system. An example of the re-learning system will be described below. FIG. 14 is a diagram showing a relearning system according to the third embodiment. The relearning system includes a plurality of information processing devices. For example, the relearning system includes information processing devices 200-205. The information processing devices 200-205 communicate with each other via a network 300. The network 300 is a wired network or a wireless network. The functions of the information processing device 100 or the information processing device 100a may be realized by a plurality of information processing devices. For example, the functions of the information processing device 100 or the information processing device 100a may be realized by the information processing devices 200-205. Therefore, the re-learning system can realize the first and second embodiments.

[0067] The features of each of the embodiments described above can be combined with each other as appropriate. [Explanation of symbols]

[0068] 10 Estimation target image, 10a region, 11 heat map, 11a range, 21-26 modified image, 21a-26a heat map, 30 modified image, 30a heat map, 40 image, 51, 52 modified image, 51a, 52a heat map, 61, 62 image, 61a, 62a image, 61a1 region, 62a1 region, 70 modified image, 70a heat map, 80 character information, 100 information processing device, 100a information processing device, 101 processor, 102 volatile storage device, 103 non-volatile storage device, 110 storage unit, 120 acquisition unit, 130 estimation unit, 140 generation unit, 140a generation unit, 150 extraction unit, 160 Output unit, 170 identification unit, 170a identification unit, 180 re-learning processing unit, 180a re-learning processing unit, 200-205 information processing device, 300 network.

Claims

1. An acquisition unit that acquires an estimation target image and a trained model; An estimation unit that performs estimation using the estimation target image and the trained model; A generation unit that generates a heat map indicating a basis for the estimation result using the estimation result; an extraction unit that extracts a plurality of features based on an area in the estimation target image, the area being a region corresponding to a location of grounds for the estimation result indicated by the heat map; A re-learning processing unit; having the generating unit, when the plurality of features are extracted, generates a plurality of modified images by performing modification for each of the features using the estimation target image; When the plurality of modified images are generated, the estimation unit performs estimation using the plurality of modified images and the trained model, When estimation is performed using the plurality of changed images and the trained model, the generation unit generates a plurality of heat maps using a plurality of estimation results, The acquisition unit acquires learning data when the estimation basis is incorrect based on estimation basis information generated based on the multiple modified images and the multiple heat maps, or the estimation target image, the heat map, the multiple modified images, and the multiple heat maps; The re-learning processing unit uses the learning data to generate re-learned data in which a feature indicated by the estimation ground is changed, and re-learns the trained model using the re-learned data. Information processing device.

2. An output unit that outputs the plurality of modified images and the plurality of heat maps, or the estimation basis information, the acquiring unit acquires the learning data when acquiring an instruction to change the feature indicated by the estimation ground and an instruction to re-learn after the plurality of modified images and the plurality of heat maps, or the estimation ground information, are output. The information processing device according to claim 1 .

3. Further comprising an identification unit that identifies, from among the plurality of heat maps, a heat map that is most different from the heat map corresponding to the estimation target image, The generation unit generates, as the estimation basis information which is character information, a feature which has been changed when a changed image corresponding to the specified heat map is generated; The acquisition unit acquires correct answer information indicating a correct estimation basis when the estimation basis information is generated, The re-learning processing unit determines whether or not the estimation basis is incorrect based on the estimation basis information and the correct answer information; The acquisition unit acquires the learning data when the estimation basis is incorrect, The re-learning processing unit generates the re-learned data in which the feature indicated by the estimation basis information is changed, using the learning data. The information processing device according to claim 1 .

4. Further comprising an identification unit that identifies, from among the plurality of heat maps, a plurality of heat maps that are different from the heat map corresponding to the estimation target image as a plurality of specific heat maps; The generation unit generates a plurality of features that have been changed when a plurality of changed images corresponding to the plurality of specific heat maps are generated as the estimation basis information, which is a plurality of character information; The acquisition unit acquires correct answer information indicating a correct estimation basis when the estimation basis information is generated, The re-learning processing unit determines whether or not there is an error in the estimation basis based on the estimation basis information and the correct answer information, The acquisition unit acquires the learning data when the estimation basis is incorrect, The re-learning processing unit uses the learning data to generate the re-learned data in which a feature having an erroneous estimation basis among a plurality of features indicated by the plurality of pieces of character information is changed. The information processing device according to claim 1 .

5. A re-learning system including a plurality of information processing devices, An acquisition unit that acquires an estimation target image and a trained model; An estimation unit that performs estimation using the estimation target image and the trained model; A generation unit that generates a heat map indicating a basis for the estimation result using the estimation result; an extraction unit that extracts a plurality of features based on an area in the estimation target image, the area being a region corresponding to a location of grounds for the estimation result indicated by the heat map; A re-learning processing unit; having the generating unit, when the plurality of features are extracted, generates a plurality of modified images by performing modification for each of the features using the estimation target image; When the plurality of modified images are generated, the estimation unit performs estimation using the plurality of modified images and the trained model, When estimation is performed using the plurality of changed images and the trained model, the generation unit generates a plurality of heat maps using a plurality of estimation results, The acquisition unit acquires learning data when the estimation basis is incorrect based on estimation basis information generated based on the multiple modified images and the multiple heat maps, or the estimation target image, the heat map, the multiple modified images, and the multiple heat maps; The re-learning processing unit uses the learning data to generate re-learned data in which a feature indicated by the estimation ground is changed, and re-learns the trained model using the re-learned data. Re-learning system.

6. An information processing device, Obtain the estimation target image and the trained model, Performing estimation using the estimation target image and the trained model; Using the estimation result, a heat map showing the basis of the estimation result is generated; Extracting a plurality of features based on an area in the estimation target image, the area being a region corresponding to a location of grounds for the estimation result indicated by the heat map; generating a plurality of modified images by performing a modification for each of the features using the estimation target image; Performing an estimation using the plurality of modified images and the trained model; Using the multiple estimation results, multiple heat maps are generated; If the estimation basis is incorrect based on estimation basis information generated based on the multiple modified images and the multiple heat maps, or the estimation target image, the heat map, the multiple modified images, and the multiple heat maps, acquiring learning data; Using the training data, re-training data is generated in which the characteristics indicated by the estimation grounds are changed; Retraining the trained model using the retrained data. How to relearn.

7. On the computer, Obtain the estimation target image and the trained model, Performing estimation using the estimation target image and the trained model; Using the estimation result, a heat map showing the basis of the estimation result is generated; Extracting a plurality of features based on an area in the estimation target image, the area being a region corresponding to a location of grounds for the estimation result indicated by the heat map; generating a plurality of modified images by performing a modification for each of the features using the estimation target image; Performing an estimation using the plurality of modified images and the trained model; Using the multiple estimation results, multiple heat maps are generated; If the estimation basis is incorrect based on estimation basis information generated based on the multiple modified images and the multiple heat maps, or the estimation target image, the heat map, the multiple modified images, and the multiple heat maps, acquiring learning data; Using the training data, re-training data is generated in which the characteristics indicated by the estimation grounds are changed; Retraining the trained model using the retrained data. A re-learning program that executes the process.

Images