Method and control device for configuring preprocessing of supervised learning model

Abstract

An exemplary method includes: defining a function g having a plurality of parameters a-l; placing the function g in front of a supervised learning model f, where the supervised learning model f is trained with training data TDi including instances xi and labels yi, where i is an integer; defining a function f(g(xi)) using the supervised learning model f and the function g; calculating a result Zi by calculating the function f(g(xi)); calculating a difference Li between the result Zi and the label yi; calculating a partial derivative PDi of the difference Li with respect to the plurality of parameters a-l of the function g; updating the plurality of parameters a-l of the function g according to the partial derivative PDi; defining a pseudo-preprocessing h using the plurality of parameters a-l of the function g if the difference Li converges during an iteration between the calculation of the result Zi and the update of the plurality of parameters a-l of the function g; placing the pseudo-preprocessing h in front of the supervised learning model f; and providing the training data TDi to the pseudo-preprocessing h to learn the supervised learning model f.

Description

[Technical field]

[0001] The present disclosure relates to methods and systems for configuring pre-processing used in supervised learning models, and in particular to methods and systems for evaluating multiple parameters used in the pre-processing. [Background technology]

[0002] The object detection device described in Patent Document 1 performs preprocessing on an image captured by a camera to detect an object in the image. In order to improve the object detection accuracy, the object detection device adjusts multiple parameters included in the preprocessing before performing the preprocessing.

[0003] FIG. 23 shows a conventional design of a preprocessing used for a supervised learning model. FIG. 23 shows a preprocessing PP1 designed by an engineer E (A) for a supervised learning model SLM (for 32-bit processing) using training data TD1 to TDn (n is an integer equal to or greater than 2). Here, the preprocessing PP1 includes, for example, multiple parameters that cannot be reproduced or are unknown to the engineer E (B) or other engineers, and the engineer E (B) is trying to convert or transfer the supervised learning model SLM from 32-bit processing to 64-bit processing.

[0004] However, engineer E (B) cannot reproduce the multiple parameters used in preprocessing PP1 because the multiple parameters used in preprocessing PP1 are unreproducible or unknown, and configures another preprocessing PP2 using other inappropriate multiple parameters different from the multiple parameters used in preprocessing PP1. As a result, engineer E (B) cannot convert or transfer the supervised learning model SLM from 32-bit processing to 64-bit processing. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-219804 A

[0006] Summary of the Invention

[0007] In order to solve the above problem, one embodiment of the present disclosure provides a method including: defining a function g having a plurality of parameters a-l; placing the function g in front of a supervised learning model f; the supervised learning model f is trained with training data TDi including instances xi and labels yi, where i is an integer; defining a function f(g(xi)) using the supervised learning model f and the function g; calculating a result Zi by calculating the function f(g(xi)); calculating a difference Li between the result Zi and the label yi; calculating a partial derivative PDi of the difference Li with respect to the plurality of parameters a-l of the function g; updating the plurality of parameters a-l of the function g according to the partial derivative PDi; if the difference Li converges during the iteration between the calculation of the result Zi and the update of the plurality of parameters a-l of the function g, defining a pseudo pre-processing h using the plurality of parameters a-l of the function g; placing the pseudo pre-processing h in front of the supervised learning model f; and providing the training data TDi to the pseudo pre-processing h to train the supervised learning model f.

[0008] Another aspect of the present disclosure is to provide a first supervised learning model f1 with training data TDi (i is an integer) to obtain a first plurality of combinations COMB_a1 to COMB_l1, each of which includes various values ​​for one of a plurality of parameters a to l of a first function g1 that is disposed before a first supervised learning model f1 that performs feedback to the first function g1; and provide a second supervised learning model f2 with training data TDi to obtain a first plurality of combinations COMB_a1 to COMB_l1, each of which includes various values ​​for one of a plurality of parameters a to l of a second function g2 that is disposed before a second supervised learning model f2 that performs feedback to the second function g2. and obtaining second multiple combinations COMB_a2 to COMB_l2 including the first and second supervised learning models f1 and f2, the operation of which is equivalent to each other, calculating first multiple distributions D_a1 to D_l1 for each of the first multiple combinations COMB_a1 to COMB_l1 and calculating second multiple distributions D_a2 to D_l2 for each of the second multiple combinations COMB_a2 to COMB_l2, and evaluating which of the first function g1 and the second function g2 is more robust by comparing the first multiple distributions D_a1 to D_l1 and the second multiple distributions D_a2 to D_l2, respectively.

[0009] Yet another aspect of the present disclosure provides a method including the steps of: acquiring a plurality of combinations COMB(1) to COMB(n), each including a plurality of values ​​of a plurality of parameters a to l of a function g, which is placed before a supervised learning model f that performs feedback to the function g, by providing training data TDi (n is an integer equal to or greater than 2) to the supervised learning model f; arranging the plurality of combinations COMB(1) to COMB(n) in a space having a plurality of dimensions equal in number to the number of the plurality of parameters a to l of the function g; acquiring at least one sample point SP in the space; selecting at least one of the combinations COMB(1) to COMB(n) that is closest to the at least one sample point SP, from the plurality of combinations COMB(1) to COMB(n); and assigning a plurality of values ​​included in the selected combination COMB(1) to COMB(n) from the plurality of combinations COMB(1) to COMB(n) to the plurality of parameters a to l of the function g.

[0010] Yet another aspect of the present disclosure includes a first definition unit that defines a function g having a plurality of parameters a to l, a first disposer that places the function g in front of a supervised learning model f, the supervised learning model f being trained with training data TDi including instances xi and labels yi, where i is an integer, a second definition unit that defines a function f(g(xi)) using the supervised learning model f and the function g, a first calculation unit that calculates the function f(g(xi)) to calculate a result Zi, a second calculation unit that calculates the difference Li between the result Zi and the label yi, and a multiplication unit that calculates the function g. The present invention provides an apparatus comprising: a third calculation unit that calculates a partial differential PDi of the difference Li with respect to multiple parameters a to l; an update unit that updates the multiple parameters a to l of the function g according to the partial differential PDi; a third definition unit that defines a pseudo preprocessing h using the multiple parameters a to l of the function g when the difference Li converges during the repetition of the calculation of the result Zi and the update of the multiple parameters a to l of the function g; a second disposer that places the pseudo preprocessing h before the supervised learning model f; and a trainer that provides training data TDi to the pseudo preprocessing h to train the supervised learning model f.

[0011] Yet another aspect of the present disclosure includes a first acquisition unit (i is an integer) that acquires a first plurality of combinations COMB_a1 to COMB_l1, each of which includes various values ​​for one of a plurality of parameters a to l of a first function g1 arranged before a first supervised learning model f1 that performs feedback to the first function g1, by providing training data TDi to the first supervised learning model f1, and a second plurality of combinations COMB_a2 to COMB_l2, each of which includes various values ​​for one of a plurality of parameters a to l of a second function g2 arranged before a second supervised learning model f2 that performs feedback to the second function g2, by providing training data TDi to the second supervised learning model f2. The present invention provides an apparatus including a second acquisition unit (the operation of the first supervised learning model f1 and the operation of the second supervised learning model f2 are equivalent to each other) that acquires training data TDi by providing the training data TDi, a first calculation unit that calculates first multiple distributions D_a1 to D_l1 for each of the first multiple combinations COMB_a1 to COMB_l1, a second calculation unit that calculates second multiple distributions D_a2 to D_l2 for each of the second multiple combinations COMB_a2 to COMB_l2, and an evaluation unit that compares the first multiple distributions D_a1 to D_l1 with the second multiple distributions D_a2 to D_l2 to evaluate which of the first function g1 and the second function g2 is more robust.

[0012] Yet another aspect of the present disclosure provides an apparatus including: a first acquisition unit that acquires a plurality of combinations COMB(1) to COMB(n), each of which includes a plurality of values ​​of a plurality of parameters a to l of a function g, by providing training data TDi (n is an integer of 2 or more) to the supervised learning model f, the plurality of combinations COMB(1) to COMB(n) being placed before a supervised learning model f that realizes feedback to the function g; a disposer that arranges the plurality of combinations COMB(1) to COMB(n) in a space having a plurality of dimensions equal in number to the number of the plurality of parameters a to l of the function g; a second acquisition unit that acquires at least one sample point SP in the space; a selection unit that selects at least one combination COMB(1) to COMB(n) that is closest to the at least one sample point SP from among the plurality of combinations COMB(1) to COMB(n); and an allocation unit that allocates a plurality of values ​​included in one of the selected plurality of combinations COMB(1) to COMB(n) to a plurality of parameters a to l of the function g. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing a configuration of a controlled system CS controlled by a control device CD according to the first embodiment. [Diagram 2] FIG. 2 shows the configuration of the function g according to the first embodiment. [Diagram 3] FIG. 3 shows the configuration of the training data TD1 to TDn. [Figure 4] FIG. 4 is a block diagram showing the functional configuration of the control device CD according to the first embodiment. [Diagram 5] FIG. 5 is a block diagram showing the configuration of the control device CD according to the first embodiment from the viewpoint of hardware. [Figure 6] FIG. 6 is a block diagram showing the configuration of the control device CD according to the first embodiment from the viewpoint of software. [Figure 7] FIG. 7 is a flowchart showing the operation (part 1) of the control device CD according to the first embodiment. [Figure 8] FIG. 8 is a flowchart showing the operation (part 2) of the control device CD according to the first embodiment. [Figure 9] FIG. 9 shows the configuration of the control device CD according to the first embodiment when i reaches 80. [Figure 10] FIG. 10 is a block diagram showing the configuration of a first controlled system CS1 and a second controlled system CS2 controlled by a control device CD according to the second embodiment. [Figure 11] FIG. 11 is a block diagram showing the functional configuration of the control device CD according to the second embodiment. [Figure 12] FIG. 12 is a flowchart showing the operation of the control device CD according to the second embodiment. [Figure 13] FIG. 13 shows a number of combinations COMB_a1 to COMB_l1 including various values ​​of a number of parameters a to l of the first function g1. [Figure 14] FIG. 14 shows a number of combinations COMB_a2 to COMB_l2 including various values ​​of a number of parameters a to l of the second function g2. [Figure 15] FIG. 15 shows a plurality of distributions from a distribution D_a1 for the combination COMB_a1 to a distribution D_l1 for the combination COMB_l1, and a plurality of distributions from a distribution D_a2 for the combination COMB_a2 to a distribution D_l2 for the combination COMB_l2. [Figure 16] FIG. 16 is a block diagram showing a configuration of a controlled system CS controlled by a control device CD according to the third embodiment. [Figure 17] FIG. 17 is a block diagram showing the functional configuration of the control device CD according to the third embodiment. [Figure 18] FIG. 18 is a flowchart showing the operation of the control device CD according to the third embodiment. [Figure 19] FIG. 19 shows a number of combinations COM(1) to COM(n) including various values ​​associated with a number of parameters a to l of the function g. [Figure 20] FIG. 20 shows a plurality of combinations COMB(1) to COMB(n) and sample points SP arranged in a space. [Figure 21]FIG. 21 shows a number of combinations COMB(1) to COMB(n) and unexpected sample points SP. [Figure 22] FIG. 22 shows a number of combinations COMB(1) to COMB(n) and some sample points SP1, SP2, and SP3. [Figure 23] FIG. 23 shows a conventional design of pre-processing used in supervised learning models. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] To explain the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0015] Embodiment 1 Configuration of the First Embodiment Hereinafter, the control device according to the first embodiment of the present disclosure will be described with reference to FIGS.

[0016] 1 is a block diagram showing the configuration of a controlled system CS controlled by a control device CD according to embodiment 1. The control device CD controls the configuration, operation, etc. of the controlled system SC.

[0017] As shown in Fig. 1, the controlled system CS includes a supervised learning model f and a function g. The configuration of the supervised learning model f has already been completely designed by an engineer E (A) (as shown in Fig. 23), and another engineer E (B) wants to train the operation of the supervised learning model f using training data TD1 to TDn (n is an integer equal to or greater than 2). The function g is an arbitrary function including multiple parameters a to l.

[0018] FIG. 2 shows the configuration of function g according to the first embodiment. As shown in FIG. 2, function g, i.e., g(x), is expressed using W, x, and z. As shown in FIG. 2, W is, for example, a matrix for conversion and normalization (e.g., a vector with 3 rows and 3 columns), x is an input image (e.g., a vector with 3 rows and 1 column), and z is a vector (e.g., a vector with 3 rows and 1 column). Here, as shown in FIG. 2, W is composed of a plurality of parameters a to i, and z is composed of a plurality of parameters i to l, so that the conclusion is drawn that function g, i.e., g(x), is composed of all of these parameters a to l.

[0019] 3 is a diagram showing the configuration of training data TD1 to TDn. For example, as shown in FIG. 3, training data TD1 has image x1 as a learning example and label y1. Here, label y1 is a correct answer indicating whether image x1, i.e., example x1, is correct when training a supervised learning model f. Like training data TD1, training data TD2 has image x2 and label y2, training data TD3 has image x3 and label y3, ..., training data TDn has image xn and label yn.

[0020] Fig. 4 is a block diagram showing the configuration of the control device CD according to the embodiment 1 from the functional aspect. As shown in Fig. 4, the control device CD according to the embodiment 1 includes a definition unit 11, a disposer 12, a calculation unit 13, an update unit 14, and a trainer 15. The functions will be described later with reference to the flowcharts of Figs. 7 and 8.

[0021] Fig. 5 is a block diagram showing, from the viewpoint of hardware, the configuration of the control device CD according to embodiment 1. As shown in Fig. 5, the control device CD includes an input circuit IC, a processing circuit PC, and an output circuit OC.

[0022] The processing circuit PC is a dedicated hardware, and realizes the functions of the definition unit 11, the disposer 12, the calculation unit 13, the update unit 14, and the trainer 15 shown in FIG.

[0023] The processing circuit PC is for example a single circuit, a composite circuit, a programmed processor, parallel programmed processors, an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a combination thereof.

[0024] Fig. 6 is a block diagram showing, from the software perspective, the configuration of the control device CD according to embodiment 1. As shown in Fig. 6, the control device CD includes an input circuit IC, a processor PR, a memory circuit SC, and an output circuit OC.

[0025] The processor PR is, for example, a CPU (including a central processing unit, a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, and a DSP (digital processing)). The processor PR implements the functions from the definition unit 11 to the trainer 15 shown in FIG.

[0026] The processor PR implements the above functions using software, firmware, or a combination of software and firmware. The software and firmware are written as programs and stored in the storage circuitry SC.

[0027] The processor PR realizes the above functions by reading and executing the program from the storage circuit SC. In other words, the program enables the computer to execute the procedures and methods from the definition unit 11 to the trainer 15.

[0028] Here, the memory circuit SC is, for example, a volatile or non-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 DVD (digital versatile disk), etc.

[0029] Of the functions from the definition unit 11 to the trainer 15, some functions may be implemented by the processing circuit PC, and other functions may be implemented by the processor PR.

[0030] As mentioned above, the functions in the control device CD from the definition unit 11 through the trainer 15 can be implemented using hardware, software, firmware, or a combination thereof.

[0031] Operation of the First Embodiment FIG. 7 is a flowchart showing the operation (part 1) of the control device CD according to the first embodiment.

[0032] FIG. 8 is a flowchart showing the operation (part 2) of the control device CD according to the first embodiment.

[0033] For ease of explanation and understanding, assume that the integer i can count up to 100, and the difference Li converges when the integer i reaches 80.

[0034] In step ST11, the definition unit 11 (shown in FIG. 4) defines a function g (shown in FIG. 1) using multiple parameters a to l (shown in FIG. 3). Furthermore, the definition unit 11 initializes the multiple parameters a to l. After the initialization by the definition unit 11, the disposer 12 (shown in FIG. 4) places or sets the function g before the supervised learning model f (shown in FIG. 1) as shown in FIG. 1.

[0035] In step ST12, the definition unit 11 defines a function f(g(x)) using a supervised learning model f and a function g, where the function f(g(x)) is formed in a nested form, i.e., the function g provides an output in response to an input image x (shown in FIG. 2 ), and the supervised learning model f provides an output in response to the output provided by the function g.

[0036] In step ST13, the definition section 11 initializes an integer i, that is, sets 1 to the integer i.

[0037] In step ST14, the calculation unit 13 (shown in FIG. 4) calculates the result Zi of the function f(g(xi)).

[0038] In step ST15, the calculation unit 13 calculates the difference Li between the result Zi and the label yi (shown in FIG. 3) by subtracting the label yi from the result Zi.

[0039] In step ST16, the calculation unit 13 calculates the partial differential PDi of the difference Li with respect to the multiple parameters a to l of the function g.

[0040] In step ST17, the update unit 14 (see FIG. 4) updates the multiple parameters a to l of the function g in accordance with the partial differential PDi.

[0041] As shown in Fig. 7, for example, when integer i is 1, multiple parameters a to l of function g are updated to a1 to l1, respectively. Similarly, as shown in Fig. 2, when integer i is 2, multiple parameters a to l of function g are updated to a2 to l2, respectively, when integer i is 3, multiple parameters a to l of function g are updated to a3 to l3, respectively, and when integer i is n, multiple parameters a to l of function g are updated to an to ln (not shown), respectively.

[0042] In step ST18, if it is determined that the difference Li has converged as a result of repeating steps ST14 to ST20, for example, by comparing with a predetermined threshold value, the process proceeds to step ST21. On the other hand, if it is determined that the difference Li has not converged, for example, by comparing with a predetermined value, the process proceeds to step ST19.

[0043] FIG. 9 shows the configuration of the control device CD according to the first embodiment when i reaches 80.

[0044] In step ST21, the definition unit 11 defines the pseudo preprocessing h using a plurality of parameters a to l when i reaches 80, that is, a plurality of parameters a80 to l80 as shown in Fig. 9. Furthermore, the disposer 12 replaces the function g with the pseudo preprocessing h, more specifically, as shown in Fig. 9, deletes the function g and places or installs the pseudo preprocessing h before the supervised learning model f.

[0045] In step ST22, trainer 15 (shown in FIG. 4) trains supervised learning model f using pseudo pre-processing h, more specifically, by providing training data TD1 to TDn to pseudo pre-processing h placed before the supervised learning model f, as shown in FIG. 9.

[0046] In step ST19, which follows after step ST18 returns NO, the integer i is incremented by one.

[0047] In step ST20, based on the above assumptions, the integer i is compared with 100. If the integer i is less than 100, the procedure returns to step ST14, otherwise the procedure ends.

[0048] Advantages of the First Embodiment As described above, the control device CD of the first embodiment defines the pseudo pre-processing h using a plurality of parameters a to l, more specifically, a plurality of parameters a80 to l80 that are given when the difference Li converges, that is, when the integer i reaches 80. This makes it possible to train the supervised learning model f using the pseudo pre-processing h instead of the function g, even if the plurality of parameters a to l of the function g are unknown or cannot be reproduced.

[0049] Embodiment 2 Configuration of the second embodiment Next, a control device according to a second embodiment of the present disclosure will be described with reference to FIGS.

[0050] 10 is a block diagram showing the configurations of a first controlled system CS1 and a second controlled system CS2 controlled by a control device CD according to embodiment 2. The control device CD according to embodiment 2 controls, for example, the configurations and operations of both the first controlled system CS1 and the second controlled system CS2, similar to the control device CD according to embodiment 1.

[0051] As shown in FIG. 10, the first controlled system CS1 includes a first supervised learning model f1 and a first function g1, and the second controlled system CS2 includes a second supervised learning model f2 and a second function g2. The operation of the first supervised learning model f1 and the operation of the second supervised learning model f2 are equivalent to each other. In the first controlled system CS1, the first function g1 is placed or installed before the first supervised learning model f1. Similarly, in the second controlled system CS2, the second function g2 is placed or installed before the second supervised learning model f2. In the second embodiment, an engineer E (B) wants to use the training data TD1 to TDn to evaluate which of the first function g1 and the second function g2 is more robust. The evaluation of the robustness of the first function g1 and the second function g2 will be described in detail later.

[0052] Fig. 11 is a block diagram showing the configuration of the control device CD according to the embodiment 2 from the functional aspect. As shown in Fig. 11, the control device CD according to the embodiment 2 includes an acquisition unit 21, a calculation unit 22, and an evaluation unit 23. The functions will be described later with reference to the flowchart of Fig. 12.

[0053] The control device CD of the second embodiment includes an input circuit IC, a processing circuit PC, and an output circuit OC (shown in FIG. 5) like the control device CD of the first embodiment. Alternatively, the control device CD according to the second embodiment may include an input circuit IC, a processor PR, a memory circuit SC, and an output circuit OC (shown in FIG. 6).

[0054] Instead of the functions of the definition unit 11, disposer 12, calculation unit 13, update unit 14, and trainer 15 of embodiment 1 (see Figure 4), a processing circuit PC (see Figure 5) or a processor PR (see Figure 6) realizes the functions of an acquisition unit 21, a calculation unit 22, and an evaluation unit 23 (see Figure 11).

[0055] Operation of the Second Embodiment FIG. 12 is a flowchart showing the operation of the control device CD according to the second embodiment.

[0056] FIG. 13 shows a number of combinations COMB_a1 to COMB_l1 including various values ​​associated with a number of parameters a to l of the first function g1.

[0057] In step ST31, the acquisition unit 21 (shown in FIG. 11) acquires a plurality of combinations COMB_a1 to COMB_l1 shown in FIG. 13. Each of the plurality of combinations COMB_a1 to COMB_l1 includes various values ​​related to one of the plurality of parameters a to l of the first function g1 (shown in FIG. 10).

[0058] For example, as shown in Fig. 13, the combination COMB_a1 includes various values ​​a1(1), a1(2), a1(3), ..., a1(n) for the parameter a of the first function g1, where the values ​​a1(1), a1(2), a1(3), ..., a1(n) are the outputs returned or fed back by the first supervised learning model f1 in response to the training data TD1, TD2, TD3, ..., TD(n).

[0059] For example, the value a1(1) is the output returned by the first supervised learning model f1 in response to training data TD1, the value a1(2) is the output returned by the first supervised learning model f1 in response to training data TD2, ..., the value a1(n) is the output returned by the first supervised learning model f1 in response to training data TDn.

[0060] FIG. 14 shows a number of combinations COMB_a2 to COMB_l2 including various values ​​of a number of parameters a to l of the second function g2.

[0061] In step ST32, the acquisition section 21 acquires a plurality of combinations COMB_a2 to COMB_l2. Each of the plurality of combinations COMB_a2 to COMB_l2 includes various values ​​for one of the plurality of parameters a to l of the second function g2.

[0062] For example, as shown in Fig. 14, the combination COMB_a2 includes various values ​​a2(1), a2(2), a2(3), ..., a2(n) for the parameter a of the second function g2, where the values ​​a2(1), a2(2), a2(3), ..., a2(n) are the outputs returned or fed back by the second supervised learning model f2 in response to the training data TD1, TD2, TD3, ..., TD(n).

[0063] For example, the value a2(1) is the output returned by the second supervised learning model f2 in response to training data TD1, the value a2(2) is the output returned by the second supervised learning model f2 in response to training data TD2, ..., the value a2(n) is the output returned by the second supervised learning model f2 in response to training data TDn.

[0064] FIG. 15 shows a plurality of distributions from a distribution D_a1 for the combination COMB_a1 to a distribution D_l1 for the combination COMB_l1, and a plurality of distributions from a distribution D_a2 for the combination COMB_a2 to a distribution D_l2 for the combination COMB_l2.

[0065] In step ST33, the calculation unit 22 (shown in FIG. 11) calculates a plurality of distributions D_l1 (shown in FIG. 15) of the combination COMB_l1 (shown in FIG. 13) from a plurality of distributions D_a1 (shown in FIG. 15) of the combination COMB_a1 (shown in FIG. 13). More specifically, the calculation unit 22 calculates the distribution D_a1 of the combination COMB_a1 including various values ​​a1(1), a1(2), a1(3), . . . , a1(n) (shown in FIG. 13) related to the parameter a of the first function g1 (shown in FIG. 10), for example. As shown in FIG. 15, the distribution D_a1 of the combination COMB_a1 has a spread peculiar to the distribution D_a1.

[0066] The calculation unit 22 calculates the other distributions D_b1 to D_l1 in the same manner as when calculating the distribution D_a1. Each of the distributions D_b1 to D_l1 has a unique extent or spread.

[0067] In step ST34, the calculation unit 22 calculates a plurality of distributions D_l2 (shown in FIG. 15) of the combination COMB_l2 (shown in FIG. 14) from a plurality of distributions D_a2 (shown in FIG. 15) of the combination COMB_a2 (shown in FIG. 14). More specifically, the calculation unit 22 calculates a distribution D_a2 of the combination COMB_a2 including various values ​​a2(1), a2(2), a2(3), . . . , a2(n) (shown in FIG. 14) associated with the parameter a of the second function g2 (shown in FIG. 10). As shown in FIG. 15, the distribution D_a2 of the combination COMB_a2 has a spread that is specific to the distribution D_a2.

[0068] Similar to the calculation of the distribution D_a2, the calculation unit 22 calculates the other distributions D_b2 to D_l2. Each of the distributions D_b2 to D_l2 has a unique extent or spread.

[0069] In step ST35, the evaluation unit 23 (shown in FIG. 11) evaluates which of the first function g1 (shown in FIG. 10) or the second function g2 (shown in FIG. 10) is more robust by comparing the spread of the multiple distributions D_a1 to D_l1 (shown in FIG. 15) with the spread of the multiple distributions D_a2 to D_l2 (shown in FIG. 15).

[0070] For example, the evaluation unit 23 compares the extent or spread of the distribution D_a1 with the extent or spread of the distribution D_a2. As shown in Fig. 15, the distribution D_a2 is wider than the distribution D_a1, so the evaluation unit 23 evaluates that the second function g2 is more robust than the first function g1 from the viewpoint of the distributions D_a1 and D_a2.

[0071] Similar to the spread of distributions D_a1 and D_a2, the evaluation unit 23 compares the spread of distributions D_b1 and D_b2, D_c1 and D_c2, . . . and D_l1 and D_l2.

[0072] The evaluation unit 23 evaluates which of the first function g1 and the second function g2 is more robust, taking into consideration all of the results of comparing the distributions D_a1 and D_a2, the distributions D_b1 and D_b2, the distributions D_c1 and D_c3, ..., and the distributions D_l1 and D_l2.

[0073] Advantages of the Second Embodiment As described above, the control device CD of embodiment 2 can evaluate which of the first function g1 and the second function g2 is more robust by comparing the extent or spread of the distributions D_a1 to D_l1 with the extent or spread of the distributions D_a2 to D_l2, respectively.

[0074] Embodiment 3 Configuration of the third embodiment Next, a control device according to a third embodiment of the present disclosure will be described with reference to FIGS.

[0075] 16 is a block diagram showing a configuration of a controlled system CS controlled by a control device CD according to embodiment 3. The control device CD according to embodiment 3 controls, for example, both the configuration and the operation of the controlled system CS, similarly to the control device CD according to embodiment 1.

[0076] As shown in Fig. 16, a controlled system CS includes a supervised learning model f and a function g, and the function g is placed or installed in front of the supervised learning model f. In the third embodiment, an engineer E (B) wants to assign various values ​​to multiple parameters a to l of the function g using training data TD1 to TDn. The above assignments will be described in detail later.

[0077] Fig. 17 is a block diagram showing the configuration of the control device CD according to the embodiment 3 from the functional aspect. As shown in Fig. 17, the control device CD according to the embodiment 3 includes an acquisition unit 31, a disposer 32, a selection unit 33, and an allocation unit 34. The functions will be described later with reference to the flowchart of Fig. 18.

[0078] The control device CD includes an input circuit IC, a processing circuit PC, and an output circuit OC (shown in FIG. 5) similar to the control device CD of the first embodiment. Alternatively, the control device CD according to the third embodiment may include an input circuit IC, a processor PR, a memory circuit SC, and an output circuit OC (shown in FIG. 6).

[0079] Instead of the functions of the definition unit 11, disposer 12, calculation unit 13, update unit 14, and trainer 15 of embodiment 1 (see Figure 4), a processing circuit PC (see Figure 5) or a processor PR (see Figure 6) realizes the functions of an acquisition unit 31, disposer 32, selection unit 33, and allocation unit 34 (see Figure 17).

[0080] Operation of the Third Embodiment FIG. 18 is a flowchart showing the operation of the control device CD according to the third embodiment.

[0081] FIG. 19 shows a number of combinations COM(1) to COM(n) including various values ​​associated with a number of parameters a to l of the function g.

[0082] In step ST41, the acquisition unit 31 (shown in FIG. 17) acquires a plurality of combinations COMB(1) to COMB(n) (shown in FIG. 19). Each of the plurality of combinations COMB(1) to COMB(n) includes various values ​​related to a plurality of parameters a to l of the function g.

[0083] For example, as shown in Fig. 19, combination COMB(1) includes various values ​​a(1), b(1), c(1), ..., l(1) associated with multiple parameters a-l of function g, where the values ​​a(1), b(1), c(1), ..., l(1) are outputs returned or fed back by supervised learning model f in response to training data TD1, the values ​​a(2), b(2), c(2), ..., l(2) are outputs returned by supervised learning model f in response to training data TD2, ..., the values ​​a(n), b(n), c(n), ..., l(n) are outputs returned by supervised learning model f in response to training data TDn.

[0084] FIG. 20 shows a plurality of combinations COMB(1) to COMB(n) and sample points SP arranged in a space.

[0085] In step ST42, the disposer 42 (shown in FIG. 17) discards or arranges in the 12-dimensional space a number of combinations COMB(1) to COMB(n) equal to the number of parameters a to l of the function g, as shown in FIG.

[0086] In step ST43, the acquisition section 41 acquires sample points SP using a known method or system, for example as shown in FIG.

[0087] In step ST44, the selection unit 33 (shown in FIG. 17) selects the combination COMB(3) that is closest to the sample point SP from among the multiple combinations COMB(1) to COMB(n) shown in FIG.

[0088] In step ST45, the assignment unit 34 (shown in FIG. 17) assigns the selected combination COMB(3), more specifically, the various values ​​a(3), b(3), c(3), ..., l(3) (shown in FIG. 19) included in the selected combination COMB(3), to multiple parameters a to l of the function g.

[0089] Advantages of the Third Embodiment As described above, the control device CD of the third embodiment selects, from among the multiple combinations COMB(1) to COMB(n), the combination COMB(3) that is located at a position closest to the sample point SP, and more specifically, selects the combination COMB(3) that can assign various values ​​a(3), b(3), c(3), ..., l(3) that may be appropriate to the multiple parameters a to l of the function g.

[0090] Modification of the third embodiment FIG. 21 shows a number of combinations COMB(1) to COMB(n) and unexpected sample points SP.

[0091] The above well-known scheme or method used for the sample points SP may rather give unexpected sample points USP (shown in FIG. 21) that are located further or furthest from some combination groups, e.g., COMB(3), COMB(6), COMB(2), and COMB(1), instead of the sample points SP (shown in FIG. 20) expected for the above groups. If only unexpected sample points USP are available for selection in the above step ST44, the assignment in step ST45 cannot give suitable values ​​for the multiple parameters a to l of the function g.

[0092] FIG. 22 shows a number of combinations COMB(1) to COMB(n) and sample points SP1, SP2, and SP3.

[0093] In order to improve the accuracy of both the above selection and assignment, in a modification of the third embodiment, as shown in FIG. 22, the acquisition unit 31 (shown in FIG. 17) acquires multiple sample points, such as three sample points SP1, SP2, and SP3, instead of one sample point SP in the third embodiment (shown in FIG. 20). Even if the sample point SP2 is unexpected, at least one of the other sample points SP1 and SP3 may be predictable, so that the selection unit 33 can select the combination COMB(3) arranged or placed closest to the sample point SP1, or the combination COMB(1) arranged or placed closest to the sample point SP3. This allows various values ​​included in the combination COMB(3) or COMB(1) that may be appropriate for the multiple parameters a to l of the function g to be assigned, as in the third embodiment.

Claims

1. Defining a function g having a number of parameters a through l; placing the function g in front of a supervised learning model f, where the supervised learning model f is trained with training data TDi that includes examples x and labels yi, where i is an integer; Defining a function f(g(xi)) using a supervised learning model f and a function g; Calculating a function f(g(xi)) to calculate a result Zi; Calculating the difference Li between the result Zi and the label yi; Calculating partial derivatives PDi of the difference Li with respect to a number of parameters a to l of the function g; updating a plurality of parameters a to l of the function g according to the partial differential PDi; defining a pseudo pre-processing h using the parameters a to l of the function g if the difference Li converges during the iterations between the calculation of the result Zi and the update of the parameters a to l of the function g; placing a pseudo pre-processing step h before the supervised learning model f; providing the training data TDi to a pseudo-preprocessing h to train a supervised learning model f; A method for providing the above.

2. obtaining a first plurality of combinations COMB_a1 to COMB_l1, each of which includes various values ​​for one of a plurality of parameters a to l of a first function g1, which is placed in front of a first supervised learning model f1 that performs feedback to the first function g1, by providing training data TDi (i is an integer) to the first supervised learning model f1; providing training data TDi to a second supervised learning model f2 to obtain a second plurality of combinations COMB_a2 to COMB_l2 each including various values ​​for one of a plurality of parameters a to l of the second function g2 placed before the second supervised learning model f2 performing feedback to the second function g2, where the operation of the first supervised learning model f1 and the operation of the second supervised learning model f2 are equivalent to each other; calculating a first plurality of distributions D_a1 to D_l1 for each of the first plurality of combinations COMB_a1 to COMB_l1; calculating a second plurality of distributions D_a2 to D_l2 for each of the second plurality of combinations COMB_a2 to COMB_l2; evaluating which of the first function g1 and the second function g2 is more robust by comparing the first plurality of distributions D_a1 to D_l1 with the second plurality of distributions D_a2 to D_l2, respectively; A method for providing the above.

3. a step of obtaining a plurality of combinations COMB(1) to COMB(n) each including a plurality of values ​​of a plurality of parameters a to l of a function g placed before a supervised learning model f that performs feedback to the function g, by providing training data TDi to the supervised learning model f, where n is an integer equal to or greater than 2; Arranging a plurality of combinations COMB(1) to COMB(n) in a space having a number of dimensions equal to the number of a plurality of parameters a to l of the function g; obtaining at least one sample point SP in a space; selecting at least one combination from among a plurality of combinations COMB(1) to COMB(n) that is closest to at least one sample point SP; assigning a plurality of values ​​included in a selected combination of a plurality of combinations COMB(1) to COMB(n) to a plurality of parameters a to l of a function g; A method for providing the above.

4. a first definition section that defines a function g having a plurality of parameters a to l; a first disposer that places a function g in front of a supervised learning model f, where the supervised learning model f is trained with training data TDi that includes examples x and labels yi, where i is an integer; A second definition section that defines a function f(g(xi)) using a supervised learning model f and a function g; a first calculation unit for calculating a function f(g(xi)) to calculate a result Zi; a second calculation unit for calculating the difference Li between the result Zi and the label yi; A third calculation unit that calculates a partial differential PDi of the difference Li with respect to a plurality of parameters a to l of the function g; an update unit that updates a plurality of parameters a to l of the function g in accordance with the partial differential PDi; a third definition part for defining a pseudo pre-processing h using the parameters a to l of the function g when the difference Li converges during the iteration between the calculation of the result Zi and the update of the parameters a to l of the function g; A second disposer that places a pseudo pre-processing h before the supervised learning model f; A trainer that provides training data TDi to a pseudo pre-processing h to train a supervised learning model f.

5. a first acquisition unit that acquires a first plurality of combinations COMB_a1 to COMB_l1, each of which includes various values ​​for one of a plurality of parameters a to l of a first function g1, which is placed in front of a first supervised learning model f1 that performs feedback to the first function g1, by providing training data TDi to the first supervised learning model f1, where i is an integer; a second acquisition unit that acquires a second plurality of combinations COMB_a2 to COMB_l2 including various values ​​for one of a plurality of parameters a to l of the second function g2 arranged before the second supervised learning model f2 that executes feedback to the second function g2 by providing training data TDi to the second supervised learning model f2, where the operation of the first supervised learning model f1 and the operation of the second supervised learning model f2 are equivalent to each other; a first calculation unit that calculates a first plurality of distributions D_a1 to D_l1 for each of the first plurality of combinations COMB_a1 to COMB_l1; a second calculation unit that calculates a second plurality of distributions D_a2 to D_l2 for each of the second plurality of combinations COMB_a2 to COMB_l2; and an evaluation unit that compares a first plurality of distributions D_a1 to D_l1 with a second plurality of distributions D_a2 to D_l2, and evaluates which of a first function g1 and a second function g2 is more robust.

6. a first acquisition unit that acquires a plurality of combinations COMB(1) to COMB(n), each of which includes a plurality of values ​​of a plurality of parameters a to l of a function g that is placed before a supervised learning model f that performs feedback to the function g, by providing training data TDi to the supervised learning model f, where n is an integer equal to or greater than 2; A disposer that arranges a plurality of combinations COMB(1) to COMB(n) in a space having a number of dimensions equal to the number of parameters a to l of the function g; A second acquisition unit that acquires at least one sample point SP in the space; a selection unit that selects at least one combination that is closest to at least one sample point SP from among the multiple combinations COMB(1) to COMB(n); an assignment unit that assigns a plurality of values ​​included in a combination selected from the plurality of combinations COMB(1) to COMB(n) to a plurality of parameters a to l of the function g.

Images