Machine learning device, machine learning method, and machine learning program

Abstract

The machine learning device is configured to include an image set acquisition unit (1) that acquires an image set including one or more images, and an image set selection unit (6) that selects an image set similar to the image set acquired by the image set acquisition unit (1) from among a plurality of image sets different from the image set acquired by the image set acquisition unit (1). The machine learning device also includes a performance comparison unit (10) that acquires performance of a learning model in which a plurality of different preprocessing processes are performed on the image sets selected by the image set selection unit (6) and machine learning is performed using the image sets after each preprocessing as performance of a plurality of machine-learned learning models, and compares the performances of the plurality of machine-learned learning models with each other, and a preprocessing acquisition unit (11) that selects one of the plurality of machine-learned learning models based on the performance comparison result by the performance comparison unit (10), and acquires the preprocessing performed on the image set used for the machine learning of the selected learning model. The machine learning device also includes a model learning unit (12) that performs preprocessing acquired by a preprocessing acquisition unit (11) on the image set acquired by the image set acquisition unit (1), and performs machine learning on an untrained learning model using the preprocessed image set.

Description

[Technical Field]

[0001] The present disclosure relates to a machine learning device, a machine learning method, and a machine learning program. [Background technology]

[0002] There is a machine learning device that uses multiple data sets to train a learning model. As such a machine learning device, for example, Patent Document 1 discloses a machine learning device equipped with a learning unit. The learning unit uses an image set including a plurality of images to train the learning model. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2023-30540 Summary of the Invention [Problem to be solved by the invention]

[0004] Images included in an image set used for machine learning of a learning model may include, for example, missing values ​​or outliers. The machine learning device disclosed in Patent Document 1 had a problem in that the performance of the learning model after machine learning by the learning unit could be reduced because the images included in the image set used for machine learning contained missing values, etc.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a machine learning device that can suppress performance degradation of a learning model after machine learning even if one or more images included in an image set contain missing values, etc. [Means for solving the problem]

[0006] A machine learning device according to the present disclosure includes an image set acquisition unit that acquires an image set including one or more images, and an image set selection unit that selects an image set similar to the image set acquired by the image set acquisition unit from among a plurality of image sets different from the image set acquired by the image set acquisition unit. The machine learning device also includes a performance comparison unit that acquires, as performance of a plurality of machine-learned learning models, the performance of a learning model in which each of a plurality of different preprocessing processes is performed on the image set selected by the image set selection unit and machine learning is performed using the image sets after each preprocessing, and compares the performance of the plurality of machine-learned learning models with each other, and a preprocessing acquisition unit that selects one of the plurality of machine-learned learning models based on the performance comparison result by the performance comparison unit, and acquires the preprocessing performed on the image set used in machine learning of the selected learning model. The machine learning device also includes a model training unit that performs preprocessing acquired by the preprocessing acquisition unit on the image set acquired by the image set acquisition unit, and causes an untrained learning model to perform machine learning using the preprocessed image set. [Effects of the Invention]

[0007] According to the present disclosure, even if one or more images included in an image set contain missing values, etc., it is possible to suppress performance degradation of a learning model after machine learning. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a configuration diagram showing a machine learning device according to a first embodiment. [Figure 2] FIG. 2 is a hardware configuration diagram showing hardware of the machine learning device according to the first embodiment. [Figure 3] FIG. 1 is a hardware configuration diagram of a computer in the case where the machine learning device is realized by software, firmware, or the like. [Figure 4] 1 is a flowchart showing a machine learning method, which is a processing procedure of the machine learning device. [Figure 5]FIG. 10 is a configuration diagram showing a machine learning device according to a second embodiment. [Figure 6] FIG. 10 is a hardware configuration diagram showing hardware of a machine learning device according to a second embodiment. [Figure 7] FIG. 10 is an explanatory diagram showing an example of a second learning model to which an image set GS″ after data dimension compression is given. DETAILED DESCRIPTION OF THE INVENTION

[0009] In order to explain the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0010] Embodiment 1 FIG. 1 is a configuration diagram showing a machine learning device according to the first embodiment. FIG. 1 is a hardware configuration diagram showing hardware of a machine learning device according to the first embodiment. The machine learning device shown in FIG. 1 includes an image set acquisition unit 1, a data storage unit 2, an image set selection unit 6, a preprocessing selection unit 9, a model learning unit 12, a model evaluation unit 13, and a model output unit 14.

[0011] The image set acquisition unit 1 is realized by, for example, an image set acquisition circuit 21 shown in FIG. The image set acquisition unit 1 acquires an image set GS containing one or more images from the outside. The image set acquisition unit 1 outputs the image set GS to each of the image set selection unit 6 and the model learning unit 12.

[0012] The data storage unit 2 is realized by, for example, a data storage circuit 22 shown in FIG. The data storage unit 2 includes a feature amount recording unit 3, a performance recording unit 4, and a preprocessing recording unit 5. The feature amount recording unit 3 records N image sets GS1 to GS2 that are different from the image set acquired by the image set acquisition unit 1. N The features Fv1 to Fv N N is an integer equal to or greater than 1.

[0013] The performance recording unit 4 records M preprocessing steps P1 to P M Among them, any of the different pretreatments P m (m=1, ,M) is performed on an N × M image set GS 1,1 ~GS N,M Each of the N × M learning models MDL 1,1 ~MDL N,M Performance S 1,1 ~S N,M M is an integer of 2 or greater. The pre-processing recording unit 5 performs M pre-processing P1 to P M It has been recorded.

[0014] The image set selection unit 6 is realized by, for example, the image set selection circuit 23 shown in FIG. The image set selection unit 6 includes a feature extraction unit 7 and a feature comparison unit 8. The image set selection unit 6 acquires the image set GS from the image set acquisition unit 1 . The image set selection unit 6 selects N image sets GS1 to GS N Among them, image set GS that is similar to image set GS SEL Select . The image set selection unit 6 selects the selected image set GS SEL is output to the preprocessing selection unit 9.

[0015] The feature extraction unit 7 acquires the image set GS from the image set acquisition unit 1 . The feature extraction unit 7 extracts the feature Fv of the image set GS, and outputs the feature Fv to the feature comparison unit 8 and the model evaluation unit 13, respectively. The feature comparison unit 8 acquires the feature Fv of the image set GS from the feature extraction unit 7 . The feature comparison unit 8 compares the feature Fv of the image set GS with the N image sets GS1 to GS2 recorded in the feature recording unit 3. N The features Fv1 to Fv N Compare with. The feature comparison unit 8 compares the feature Fv of the image set GS with the N feature Fv1 to Fv N Based on the comparison results, N image sets GS1 to GS N image set GS similar to the image set GS acquired by the image set acquisition unit 1 SEL Select . The feature comparison unit 8 compares the selected image set GS SEL is output to the performance comparison unit 10.

[0016] The pre-processing selection unit 9 is realized by, for example, the pre-processing selection circuit 24 shown in FIG. The pre-processing selection unit 9 includes a performance comparison unit 10 and a pre-processing acquisition unit 11. The pre-processing selection unit 9 selects M pre-processings P1 to P2 recorded in the pre-processing recording unit 5. M Among these, a pre-processing P suitable for the image set GS acquired by the image set acquisition unit 1 is selected. SEL Get. The pre-processing selection unit 9 selects the pre-processing P SEL is output to the model learning unit 12.

[0017] The performance comparison unit 10 compares the image set GS selected by the image set selection unit 6. SEL Get. The performance comparison unit 10 selects M machine-learned learning models MDL SEL,1 ~MDL SEL,M Performance S SEL,1 ~S SEL,M The image set GS selected by the image set selection unit 6 is SEL For each, M different preprocessing P1 to P M Each of the preprocessed image sets GS is SEL,m The learning model MDL was machine-learned using (m=1, ,M) SEL,m Performance S SEL,m Get. The performance comparison unit 10 compares M performance S SEL,1 ~Performance S SEL,M The performance comparison results are output to the preprocessing acquisition unit 11.

[0018] The preprocessing acquisition unit 11 acquires the performance comparison results from the performance comparison unit 10 . The preprocessing acquisition unit 11 generates M learning models MDL based on the performance comparison results by the performance comparison unit . SEL,1 ~MDL SEL,M From among these, one of the learning models MDL SEL Select . The preprocessing acquisition unit 11 acquires M preprocessing operations P1 to P2 recorded in the preprocessing recording unit 5. M Among them, pre-processing P SEL As a learning model, MDL SEL Image set GS used for machine learning SEL,m Pretreatment P m Get. The pre-processing acquisition unit 11 acquires the pre-processing P SEL are output to the model learning unit 12 and the model evaluation unit 13, respectively.

[0019] The model learning unit 12 is realized by, for example, a model learning circuit 25 shown in FIG. The model learning unit 12 acquires the image set GS from the image set acquisition unit 1, and acquires the preprocessing P SEL Get. The model learning unit 12 performs preprocessing P SEL Then, the preprocessed image set GS' is used to perform machine learning on the untrained learning model MDL. The model learning unit 12 outputs the learned model MDL′ after machine learning to the model evaluation unit 13. The model learning unit 12 selects a plurality of preprocessing P SEL If is selected, for the image set GS, each preprocessing P SEL Then, the untrained learning model MDL is used to perform machine learning using the image set GS' after each preprocessing. The model learning unit 12 outputs a plurality of machine-learned learning models MDL′ to the model evaluation unit 13.

[0020] The model evaluation unit 13 is realized by, for example, a model evaluation circuit 26 shown in FIG. The model evaluation unit 13 acquires the machine-learned learning model MDL′ from the model learning unit 12. The model evaluation unit 13 outputs the machine-learned learning model MDL′ to the model output unit 14. The model evaluation unit 13 evaluates the performance of the learning model MDL′ after machine learning, and causes the performance recording unit 4 to record the performance of the learning model MDL′. Furthermore, the model evaluation unit 13 records the feature Fv extracted by the feature extraction unit 7 in the feature recording unit 3, and the preprocessing P acquired by the preprocessing acquisition unit 11. SEL is recorded in the preprocessing recording unit 5.

[0021] When the model evaluation unit 13 acquires multiple machine-learned learning models MDL' from the model learning unit 12, it evaluates the performance of each of the multiple machine-learned learning models MDL', and selects one or more learning models MDL' from among the multiple machine-learned learning models MDL' based on the performance evaluation results. The model evaluation unit 13 outputs the selected learning model MDL′ to the model output unit 14. The model evaluation unit 13 causes the performance recording unit 4 to record the performance of the selected learning model MDL′. Furthermore, the model evaluation unit 13 records the feature Fv extracted by the feature extraction unit 7 in the feature recording unit 3, and the preprocessing P acquired by the preprocessing acquisition unit 11. SEL is recorded in the preprocessing recording unit 5.

[0022] The model output unit 14 is realized by, for example, a model output circuit 27 shown in FIG. The model output unit 14 acquires the machine-learned learning model MDL′ from the model evaluation unit 13. The model output unit 14 outputs the machine-learned learning model MDL' to, for example, a device that uses the learning model MDL'.

[0023] In Figure 1, it is assumed that each of the components of the machine learning device, namely, image set acquisition unit 1, data storage unit 2, image set selection unit 6, preprocessing selection unit 9, model learning unit 12, model evaluation unit 13, and model output unit 14, is realized by dedicated hardware such as that shown in Figure 2. In other words, it is assumed that the machine learning device is realized by an image set acquisition circuit 21, data storage circuit 22, image set selection circuit 23, preprocessing selection circuit 24, model learning circuit 25, model evaluation circuit 26, and model output circuit 27.

[0024] Here, the data storage circuit 22 may be, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, or a DVD (Digital Versatile Disc). Furthermore, each of the image set acquisition circuit 21, image set selection circuit 23, preprocessing selection circuit 24, model learning circuit 25, model evaluation circuit 26, and model output circuit 27 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination thereof.

[0025] The components of the machine learning device are not limited to those realized by dedicated hardware, and the machine learning device may be realized by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the memory of a computer. A computer refers to hardware that executes the program, such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), central processing unit, processing unit, arithmetic unit, microprocessor, microcomputer, processor, or DSP (Digital Signal Processor).

[0026] FIG. 3 is a hardware configuration diagram of a computer in the case where the machine learning device is realized by software, firmware, or the like. When the machine learning device is realized by software, firmware, or the like, the data storage unit 2 is configured on the computer's memory 31. A machine learning program is stored in the memory 31 to cause the computer to execute each of the processing procedures in the image set acquisition unit 1, image set selection unit 6, performance comparison unit 10, preprocessing acquisition unit 11, model learning unit 12, model evaluation unit 13, and model output unit 14, namely, an image set acquisition procedure, an image set selection procedure, a performance comparison procedure, a preprocessing acquisition procedure, a model learning procedure, a model evaluation procedure, and a model output procedure. A processor 32 of the computer then executes the machine learning program stored in the memory 31.

[0027] 2 shows an example in which each of the components of the machine learning device is realized by dedicated hardware, while FIG. 3 shows an example in which the machine learning device is realized by software, firmware, etc. However, this is merely an example, and some of the components in the machine learning device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0028] Next, the operation of the machine learning device shown in FIG. 1 will be described. FIG. 4 is a flowchart showing a machine learning method, which is a processing procedure of the machine learning device. The image set acquisition unit 1 acquires an image set GS including one or more images from the outside (step ST1 in FIG. 4). The image set GS may be acquired by any method, for example, by reading it from an external storage device (not shown), or by reading it from a communication device (not shown) via a network. The image set acquisition unit 1 outputs the image set GS to each of the image set selection unit 6 and the model learning unit 12.

[0029] The feature extraction unit 7 of the image set selection unit 6 acquires the image set GS from the image set acquisition unit 1 . The feature extraction unit 7 extracts the feature Fv of the image set GS (step ST2 in FIG. 4). The feature Fv extracted by the feature extraction unit 7 may be, for example, the average value of the pixel values ​​of the N images included in the image set GS and the variance value of the pixel values ​​of the N images, expressed in the form of a multidimensional vector. Furthermore, the feature Fv may be, for example, a vector space in which N images included in the image set GS are embedded and each image is represented by a set of vectors. Furthermore, the feature Fv may be, for example, a feature in which N images included in the image set GS are embedded in an arbitrary vector space, and the distribution of the image set GS in the vector space is expressed. The feature extraction unit 7 outputs the feature Fv of the image set GS to the feature comparison unit 8 and the model evaluation unit 13, respectively.

[0030] The feature comparison unit 8 acquires the feature Fv of the image set GS from the feature extraction unit 7 . The feature comparison unit 8 selects N image sets GS1 to GS2 from the feature recording unit 3. N The features Fv1 to Fv N Get. The feature comparison unit 8 compares the feature Fv of the image set GS with the feature Fv of the image set GS n (n=1, ,N) feature Fvn and compare the feature Fv and feature Fv n Based on the comparison results, image set GS and image set GS n The degree to which they are similar is determined. Image set GS and image sets GS1~GS N As a method for determining the similarity between the feature Fv and the feature Fv1 to Fv2, the feature comparison unit 8 may, for example, N Euclidean distance between feature Fv and feature Fv1~Fv N The Mahalanobis distance between feature Fv and feature Fv1 to Fv N The feature comparison unit 8 then calculates the Frechet Inception Distance between each of the N Euclidean distances, the N Mahalanobis distances, or the N Frechet Inception distances. The feature comparison unit 8 can use a method of comparing the N Euclidean distances, the N Mahalanobis distances, or the N Frechet Inception distances with each other, and making a determination based on the comparison result of the Euclidean distances, the comparison result of the Mahalanobis distances, or the comparison result of the Frechet Inception distances. Also, image set GS and image sets GS1~GS N As a method for determining the similarity between the feature Fv and the feature Fv1 to Fv2, the feature comparison unit 8 may, for example, N Cosine similarity between feature Fv and feature Fv1 to Fv N Then, the feature comparison unit 8 can use a method of comparing the N cosine similarities or the N Dice coefficients with each other and making a determination based on the comparison result of the cosine similarities or the comparison result of the Dice coefficients.

[0031] The feature comparison unit 8 compares the feature Fv of the image set GS with the N feature Fv1 to Fv N Based on the comparison results, N image sets GS1 to GS N Among them, image sets GS similar to image set GS SEL (Step ST3 in FIG. 4). Specifically, the feature comparison unit 8 compares the feature Fv of the image set GS with the N feature Fv1 to Fv N Based on the comparison results, N image sets GS1 to GS NAmong them, the image set GS that is most similar to the image set GS SEL Select . Here, the feature comparison unit 8 compares N image sets GS1 to GS N Among them, the image set GS that is most similar to the image set GS SEL However, this is only an example, and the feature comparison unit 8 selects N image sets GS1 to GS N From the image set GS, the similarity is greater than or equal to the threshold. SEL For example, N image sets GS1 to GS N If the image sets GS2 and GS5 have similarities above the threshold, then the image set GS SEL The image sets GS2 and GS5 are selected as the threshold. The threshold may be stored in an internal memory of the feature comparison unit 8, or may be provided from outside the machine learning device. The feature comparison unit 8 compares the selected image set GS SEL is output to the performance comparison unit 10.

[0032] The performance comparison unit 10 receives the image set GS SEL Get. Furthermore, the performance comparison unit 10 receives M machine-learned learning models MDL SEL,1 ~MDL SEL,M Performance S SEL,1 ~S SEL,M Get. Machine-learned learning model MDL SEL,m (m=1, ,M) is the image set GS selected by the image set selector 6. SEL For the pretreatment P m Image set GS SEL,m This is a learning model that was machine-learned using Pretreatment P m Examples of preprocessing include missing value processing, outlier processing, resizing, and filtering. m If the number of images is M, the performance recording unit 4 stores the image set GS selected by the image set selection unit 6. SEL For pretreatment P m(m=1, ,M) was performed, and machine learning was performed using the preprocessed image set. SEL,m Performance S SEL,m Assuming that M performance S SEL,1 ~Performance S SEL,M has been recorded. The performance comparison unit 10 compares M performance S SEL,1 ~S SEL,M The performance comparison results are output to the preprocessing acquisition unit 11 (step ST4 in FIG. 4).

[0033] The preprocessing acquisition unit 11 acquires the performance comparison results from the performance comparison unit 10 . The preprocessing acquisition unit 11 generates M learning models MDL based on the performance comparison results by the performance comparison unit . SEL,1 ~MDL SEL,M One or more learning models MDL SEL (Step ST5 in FIG. 4). Specifically, the preprocessing acquisition unit 11 acquires M learning models MDL SEL,1 ~MDL SEL,M From among these, the learning model MDL SEL For example, the learning model with the highest performance may be selected, or the top few learning models with the highest performance may be selected, or a learning model with performance equal to or above a threshold may be selected. The preprocessing acquisition unit 11 acquires M preprocessing operations P1 to P2 recorded in the preprocessing recording unit 5. M Among them, preprocessing P suitable for image set GS is SEL As a learning model, MDL SEL Image set GS used for machine learning SEL,m Pretreatment P m (Step ST6 in FIG. 4). The preprocessing acquisition unit 11 performs preprocessing P SEL are output to the model learning unit 12 and the model evaluation unit 13, respectively.

[0034] The model learning unit 12 acquires the image set GS from the image set acquisition unit 1, and acquires the preprocessing P SEL Get. The model learning unit 12 performs preprocessing P SEL (Step ST7 in FIG. 4). The model learning unit 12 performs a plurality of preprocessing P SEL If the image set GS is obtained, each preprocessing P SEL Do the following.

[0035] The model training unit 12 uses the preprocessed image set GS' to cause the untrained training model MDL to perform machine learning (step ST8 in FIG. 4). The untrained training model MDL may be stored in an internal memory of the model training unit 12, or may be stored in a storage device or the like provided externally to the model training unit 12. The model learning unit 12 performs a plurality of preprocessing P SEL If GS′ has been obtained, the untrained learning model MDL is used to perform machine learning using the image set GS′ after each preprocessing. The model learning unit 12 outputs a plurality of machine-learned learning models MDL′ to the model evaluation unit 13.

[0036] When the model evaluation unit 13 acquires one machine-learned learning model MDL' from the model learning unit 12, the model evaluation unit 13 evaluates the performance S of the machine-learned learning model MDL' (step ST9 in FIG. 4). When the model evaluation unit 13 acquires a plurality of machine-learned learning models MDL' from the model learning unit 12, the model evaluation unit 13 evaluates the performance S of each machine-learned learning model MDL' (step ST9 in FIG. 4). Methods for evaluating the performance S include, for example, a method based on the magnitude of loss when machine learning the learning model MDL', a method based on the accuracy rate using an evaluation image set, or a method using the AUC (Area Under the Curve) for the evaluation image set as an index.

[0037] When the model evaluation unit 13 acquires one machine-learned learning model MDL′ from the model learning unit 12, the model evaluation unit 13 outputs one machine-learned learning model MDL′ to the model output unit 14. When the model evaluation unit 13 acquires multiple machine-learned learning models MDL' from the model learning unit 12, it outputs the learning model MDL' with the highest evaluation of performance S among the multiple machine-learned learning models MDL' to the model output unit 14. Here, the model evaluation unit 13 outputs the learning model MDL' with the highest evaluation of performance S among the multiple learning models MDL' after machine learning to the model output unit 14. However, this is just one example, and the model evaluation unit 13 may also output the top several learning models MDL' with the highest evaluation of performance S among the multiple learning models MDL' after machine learning to the model output unit 14. Furthermore, the model evaluation unit 13 may output to the model output unit 14, among a plurality of machine-learned learning models MDL′, a learning model MDL′ whose evaluation of performance S is equal to or greater than a threshold.

[0038] The model evaluation unit 13 causes the performance recording unit 4 to record the performance S of the learning model MDL′ output to the model output unit 14 . Furthermore, the model evaluation unit 13 records the feature Fv extracted by the feature extraction unit 7 in the feature recording unit 3, and the preprocessing P acquired by the preprocessing acquisition unit 11. SEL is recorded in the preprocessing recording unit 5.

[0039] The model output unit 14 acquires the machine-learned learning model MDL′ from the model evaluation unit 13. The model output unit 14 outputs the machine-learned learning model MDL' to, for example, a device that uses the learning model MDL'.

[0040] In the first embodiment described above, the machine learning device is configured to include an image set acquisition unit 1 that acquires an image set including one or more images, and an image set selection unit 6 that selects an image set similar to the image set acquired by the image set acquisition unit 1 from a plurality of image sets different from the image set acquired by the image set acquisition unit 1. The machine learning device also includes a performance comparison unit 10 that acquires the performance of a plurality of machine-learned learning models in which a plurality of different preprocessing processes are performed on the image set selected by the image set selection unit 6 and machine learning is performed using the image sets after each preprocessing, and compares the performance of the plurality of machine-learned learning models with each other, and a preprocessing acquisition unit 11 that selects one of the plurality of machine-learned learning models based on the performance comparison result by the performance comparison unit 10 and acquires the preprocessing performed on the image set used for machine learning of the selected learning model. The machine learning device also includes a model learning unit 12 that performs the preprocessing acquired by the preprocessing acquisition unit 11 on the image set acquired by the image set acquisition unit 1, and causes an untrained learning model to perform machine learning using the preprocessed image set. Therefore, even if one or more images included in the image set contain missing values ​​or the like, the machine learning device can suppress a decrease in the performance of the learning model after machine learning.

[0041] 1, the model learning unit 12 causes the untrained learning model MDL to perform machine learning using the preprocessed image set GS' selected by the preprocessing selection unit 9. However, this is merely one example, and the model learning unit 12 may cause the untrained learning model MDL to perform machine learning using the preprocessed image set GS' selected by the preprocessing selection unit 9, or may cause the untrained learning model MDL to perform machine learning using the image set GS acquired by the image set acquisition unit 1 without performing preprocessing on the image set GS. In this case, the model evaluation unit 13 evaluates the performance S of the learning model MDL' after machine learning using the preprocessed image set GS' selected by the preprocessing selection unit 9, and evaluates the performance S of the learning model MDL' after machine learning using the image set GS acquired by the image set acquisition unit 1. Then, based on the evaluation results of the performance S of these learning models MDL' after machine learning, the model evaluation unit 13 selects one or more learning models MDL' from these learning models MDL' after machine learning.

[0042] Embodiment 2 In the second embodiment, a machine learning device including a data dimension compression unit 15 that compresses the data dimension of the image set GS acquired by the image set acquisition unit 1 will be described.

[0043] Fig. 5 is a configuration diagram showing a machine learning device according to embodiment 2. In Fig. 5, the same reference numerals as in Fig. 1 indicate the same or corresponding parts, and detailed description thereof will be omitted. Fig. 6 is a hardware configuration diagram showing hardware of a machine learning device according to embodiment 2. In Fig. 6, the same reference numerals as in Fig. 2 indicate the same or corresponding parts, and detailed description thereof will be omitted. The machine learning device shown in Figure 5 includes an image set acquisition unit 1, a data dimension compression unit 15, a data storage unit 2, an image set selection unit 6, a preprocessing selection unit 9, a model learning unit 12, a model evaluation unit 13, and a model output unit 14.

[0044] The data dimension compression unit 15 is realized by, for example, a data dimension compression circuit 28 shown in FIG. The data dimension compression unit 15 acquires the image set GS from the image set acquisition unit 1 . The data dimension compression unit 15 compresses the data dimension of the image set GS. The data dimension compression unit 15 outputs the image set GS″ after data dimension compression to the feature amount extraction unit 7 a of the image set selection unit 6 .

[0045] In the machine learning device shown in FIG. 5, the image set selection unit 6 includes a feature extraction unit 7a and a feature comparison unit 8. The feature extraction unit 7a acquires the image set GS″ after data dimension compression from the data dimension compression unit 15. The feature extraction unit 7a provides the image set GS″ after data dimension compression to a second learning model such as that shown in FIG. The feature extraction unit 7a acquires a set of vectors output from either the intermediate layer of the second learning model or the output layer of the second learning model as a feature Fv to be extracted. The feature extraction unit 7 a outputs the feature Fv to the feature comparison unit 8 . FIG. 7 is an explanatory diagram showing an example of the second learning model to which the image set GS″ after data dimension compression is given. The second learning model is a learning model that, during learning, is given, for example, an image set after data dimension compression and training data indicating the features of the image set, and machine-learns the features of the image set. During inference, when the image set GS″ after data dimension compression is given to the input layer, the second learning model outputs the feature Fv of the image set GS″ from the intermediate layer or the output layer.

[0046] In Figure 5, it is assumed that each of the components of the machine learning device, namely, the image set acquisition unit 1, the data dimension compression unit 15, the data storage unit 2, the image set selection unit 6, the preprocessing selection unit 9, the model learning unit 12, the model evaluation unit 13, and the model output unit 14, is realized by dedicated hardware such as that shown in Figure 6. In other words, it is assumed that the machine learning device is realized by an image set acquisition circuit 21, a data dimension compression circuit 28, a data storage circuit 22, an image set selection circuit 23, a preprocessing selection circuit 24, a model learning circuit 25, a model evaluation circuit 26, and a model output circuit 27. Each of the image set acquisition circuit 21, the data dimension compression circuit 28, the image set selection circuit 23, the preprocessing selection circuit 24, the model learning circuit 25, the model evaluation circuit 26, and the model output circuit 27 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof.

[0047] The components of the machine learning device are not limited to those realized by dedicated hardware, and the machine learning device may be realized by software, firmware, or a combination of software and firmware. When the machine learning device is realized by software, firmware, or the like, the data storage unit 2 is configured on the memory 31 shown in Fig. 3. A machine learning program for causing a computer to execute each of the processing procedures in the image set acquisition unit 1, data dimension compression unit 15, image set selection unit 6, performance comparison unit 10, preprocessing acquisition unit 11, model learning unit 12, model evaluation unit 13, and model output unit 14, including an image set acquisition procedure, a data dimension compression procedure, an image set selection procedure, a performance comparison procedure, a preprocessing acquisition procedure, a model learning procedure, a model evaluation procedure, and a model output procedure, is stored in the memory 31 shown in Fig. 3. Then, the processor 32 shown in Fig. 3 executes the machine learning program stored in the memory 31.

[0048] 6 shows an example in which each of the components of the machine learning device is realized by dedicated hardware, while FIG. 3 shows an example in which the machine learning device is realized by software, firmware, etc. However, this is merely an example, and some of the components in the machine learning device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0049] Next, the operation of the machine learning device shown in FIG. 5 will be described. However, other than the data dimension compression unit 15 and the feature amount extraction unit 7a, the machine learning device is the same as that shown in Fig. 1. Therefore, only the operations of the data dimension compression unit 15 and the feature amount extraction unit 7a will be described here.

[0050] The data dimension compression unit 15 acquires the image set GS from the image set acquisition unit 1 . The data dimension compression unit 15 compresses the data dimension of the image set GS. One method for compressing data dimensions is to provide an image set GS to the input layer of a convolutional neural network (hereinafter referred to as "CNN") that has been trained using an image set for ImageNet classification tasks, and output a set of vectors output from either the intermediate layer or the output layer of the CNN as an image set GS after data dimension compression. In the machine learning device shown in FIG. 5, the image set GS″ after data dimension compression by the data dimension compression unit 15 is an image set whose data size has been reduced by an arbitrary method. The data dimension compression unit 15 outputs the image set GS″ after data dimension compression to the feature amount extraction unit 7 a of the image set selection unit 6 .

[0051] The feature extraction unit 7a acquires the image set GS″ after data dimension compression from the data dimension compression unit 15. The feature extraction unit 7a provides each image included in the image set GS″ after data dimension compression to, for example, one of the input layers in the second learning model as shown in FIG. The feature extraction unit 7a acquires a set of vectors output from either the intermediate layer of the second learning model or the output layer of the second learning model as a feature Fv to be extracted. In the example of Fig. 7, the second learning model has a three-stage intermediate layer. However, this is merely an example, and the second learning model may have two or fewer intermediate layers, or four or more intermediate layers. For example, in the case of the second learning model having a three-stage intermediate layer, a set of vectors output from one of the three intermediate layers is acquired as the feature Fv to be extracted. Furthermore, the feature extraction unit 7a may obtain a probability mass function of a vector set output from the intermediate layer of the second learning model or the output layer of the second learning model, and use the probability mass function as the feature Fv.

[0052] In the above-described second embodiment, the machine learning device shown in FIG. 5 is configured to include a data dimension compression unit 15 that compresses the data dimensions of the image set acquired by the image set acquisition unit 1. Therefore, similar to the machine learning device shown in FIG. 1, the machine learning device shown in FIG. 5 can suppress performance degradation of the learning model after machine learning and reduce unnecessary information even if one or more images included in the image set contain missing values, etc. Therefore, the machine learning device shown in FIG. 5 can improve the accuracy of feature calculation and speed up feature comparison processing compared to the machine learning device shown in FIG. 1.

[0053] In the machine learning device shown in FIG. 5, the feature extraction unit 7a obtains the feature Fv to be extracted by providing the image set GS″ after data dimension compression to the input layer of the second learning model. However, this is merely an example, and the feature extraction unit 7a may be configured to extract the feature Fv of the image set GS″ in the same way as the feature extraction unit 7 shown in FIG. 1.

[0054] In the machine learning device shown in FIG. 5, the feature extraction unit 7a obtains the feature Fv to be extracted by providing the image set GS″ after data dimension compression to the input layer of the second learning model. However, this is merely an example, and the feature extraction unit 7a may provide the image set GS obtained by the image set obtaining unit 1 to the input layer of the second learning model, and obtain a vector set output from either the intermediate layer of the second learning model or the output layer of the second learning model as the feature Fv to be extracted. In this case, the second learning model is a learning model that, during learning, is given, for example, an image set and training data indicating the features of the image set, and machine-learns the features of the image set. During inference, when an image set GS is given to the input layer, the second learning model outputs the feature Fv of the image set GS from the intermediate layer or the output layer.

[0055] In addition, the present disclosure allows for free combination of the respective embodiments, modification of any of the components of the respective embodiments, or omission of any of the components of the respective embodiments. [Industrial Applicability]

[0056] The present disclosure is suitable for a machine learning device, a machine learning method, and a machine learning program. [Explanation of symbols]

[0057] 1 image set acquisition unit, 2 data storage unit, 3 feature recording unit, 4 performance recording unit, 5 preprocessing recording unit, 6 image set selection unit, 7 feature extraction unit, 7a feature extraction unit, 8 feature comparison unit, 9 preprocessing selection unit, 10 performance comparison unit, 11 preprocessing acquisition unit, 12 model learning unit, 13 model evaluation unit, 14 model output unit, 15 data dimension compression unit, 21 image set acquisition circuit, 22 data storage circuit, 23 image set selection circuit, 24 preprocessing selection circuit, 25 model learning circuit, 26 model evaluation circuit, 27 model output circuit, 28 data dimension compression circuit, 31 memory, 32 processor.

Claims

1. an image set acquisition unit for acquiring an image set including one or more images; an image set selection unit that selects an image set similar to the image set acquired by the image set acquisition unit from among a plurality of image sets different from the image set acquired by the image set acquisition unit; A performance comparison unit that performs a plurality of different preprocessing processes on the image sets selected by the image set selection unit as the performance of a plurality of machine-learned learning models, obtains the performance of a learning model in which machine learning is performed using the image sets after each preprocessing, and compares the performances of the plurality of machine-learned learning models with each other; and A preprocessing acquisition unit that selects a learning model from the plurality of machine-learned learning models based on a performance comparison result by the performance comparison unit, and acquires preprocessing performed on the image set used in the machine learning of the selected learning model; a model learning unit that performs preprocessing on the image set acquired by the image set acquisition unit, the preprocessing being acquired by the preprocessing acquisition unit, and causes an untrained learning model to perform machine learning using the image set after the preprocessing; A machine learning device equipped with

2. The image set selection unit a feature extraction unit that extracts features of the image set acquired by the image set acquisition unit; 2. The machine learning device according to claim 1, further comprising a feature comparison unit that compares features extracted from each of a plurality of image sets different from the image set acquired by the image set acquisition unit with features extracted by the feature extraction unit, and selects an image set similar to the image set acquired by the image set acquisition unit from among the plurality of image sets based on a comparison result of the features.

3. A plurality of preprocessing processes are acquired by the preprocessing acquisition unit, The model learning unit performs each preprocessing acquired by the preprocessing acquisition unit on the image set acquired by the image set acquisition unit, and performs machine learning on an untrained learning model using the image sets after each preprocessing, The machine learning device according to claim 1, further comprising a model evaluation unit that evaluates the performance of each of the multiple machine-learned learning models by the model learning unit, and selects one or more learning models from among the multiple machine-learned learning models based on the performance evaluation results.

4. The model learning unit includes: In addition to performing machine learning on an untrained learning model using each preprocessed image set, the image set acquired by the image set acquisition unit is not preprocessed, and the untrained learning model is performed on the image set; The model evaluation unit includes: The machine learning device according to claim 3, characterized in that the performance of each of the multiple machine-learned learning models by the model learning unit is evaluated, and one or more learning models are selected from the multiple machine-learned learning models based on the performance evaluation results.

5. The machine learning device according to claim 1, further comprising a data dimension compression unit that compresses the data dimension of the image set acquired by the image set acquisition unit and outputs the image set after data dimension compression to the image set selection unit.

6. The feature extraction unit The machine learning device according to claim 2, characterized in that the image set acquired by the image set acquisition unit is provided to a second learning model, and a set of vectors output from either an intermediate layer of the second learning model or an output layer of the second learning model is acquired as a feature to be extracted.

7. a data dimension compression unit that compresses a data dimension of the image set acquired by the image set acquisition unit, The feature extraction unit The machine learning device according to claim 2, characterized in that the image set after data dimension compression by the data dimension compression unit is provided to a second learning model, and a set of vectors output from either an intermediate layer of the second learning model or an output layer of the second learning model is obtained as a feature to be extracted.

8. The feature comparison unit is The machine learning device according to claim 2, further comprising: a Fréchet initiation distance between a feature extracted from each of a plurality of image sets different from the image set acquired by the image set acquisition unit and the feature extracted by the feature extraction unit; a comparison of the plurality of Fréchet initiation distances; and a selection of an image set similar to the image set acquired by the image set acquisition unit from among the plurality of image sets based on a comparison result of the Fréchet initiation distances.

9. an image set acquisition unit acquires an image set including one or more images; an image set selection unit selects an image set similar to the image set acquired by the image set acquisition unit from among a plurality of image sets different from the image set acquired by the image set acquisition unit; The performance comparison unit obtains, as the performance of a plurality of machine-learned learning models, the performance of a learning model in which a plurality of different preprocessing processes are performed on the image set selected by the image set selection unit, and machine learning is performed using the image set after each preprocessing, and compares the performances of the plurality of machine-learned learning models with each other; a preprocessing acquisition unit that selects a learning model from among the plurality of machine-learned learning models based on a performance comparison result by the performance comparison unit, and acquires preprocessing performed on the image set used in the machine learning of the selected learning model; A model learning unit performs the preprocessing acquired by the preprocessing acquisition unit on the image set acquired by the image set acquisition unit, and causes an untrained learning model to perform machine learning using the image set after the preprocessing. Machine learning methods.

10. an image set acquisition step in which an image set acquisition unit acquires an image set including one or more images; an image set selection step in which an image set selection unit selects an image set similar to the image set acquired by the image set acquisition unit from among a plurality of image sets different from the image set acquired by the image set acquisition unit; a performance comparison step in which a performance comparison unit obtains performance of a learning model in which a plurality of different preprocessing processes are performed on the image set selected by the image set selection unit, and machine learning is performed using the image set after each preprocessing, as performance of a plurality of machine-learned learning models, and compares the performances of the plurality of machine-learned learning models with each other; a preprocessing acquisition step in which a preprocessing acquisition unit selects one of the multiple machine-learned learning models based on a performance comparison result by the performance comparison unit, and acquires preprocessing performed on the image set used in the machine learning of the selected learning model; a model learning procedure in which a model learning unit performs preprocessing on the image set acquired by the image set acquisition unit, the preprocessing being acquired by the preprocessing acquisition unit, and causes an untrained learning model to perform machine learning using the image set after the preprocessing; A machine learning program for a computer to run.

Images