WO2022113338A1 - 情報処理装置、情報処理方法、及び、記録媒体 - Google Patents

情報処理装置、情報処理方法、及び、記録媒体 Download PDF

Info

Publication number
WO2022113338A1
WO2022113338A1 PCT/JP2020/044483 JP2020044483W WO2022113338A1 WO 2022113338 A1 WO2022113338 A1 WO 2022113338A1 JP 2020044483 W JP2020044483 W JP 2020044483W WO 2022113338 A1 WO2022113338 A1 WO 2022113338A1
Authority
WO
WIPO (PCT)
Prior art keywords
error
model
prediction
label
student
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2020/044483
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
優太 畠山
穣 岡嶋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NEC Corp
Original Assignee
NEC Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by NEC Corp filed Critical NEC Corp
Priority to JP2022564995A priority Critical patent/JP7529040B2/ja
Priority to PCT/JP2020/044483 priority patent/WO2022113338A1/ja
Priority to US18/037,298 priority patent/US20240020575A1/en
Publication of WO2022113338A1 publication Critical patent/WO2022113338A1/ja
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/01—Probabilistic graphical models, e.g. probabilistic networks

Definitions

  • the present invention relates to a technique for improving the accuracy of a machine learning model.
  • Active learning is known as a technique for improving the accuracy of machine learning models through supervised learning. Active learning is the accuracy of a model by a teacher (Oracle) assigning a label to an example that cannot be predicted well with the current machine learning model, generating an example, and retraining the machine learning model using the example. It is a method to improve.
  • the active learning method basically regards "an example in which the student model outputs ambiguous or inconsistent predictions" as an example that cannot be predicted well, and labels the example and retrains it.
  • Uncertainty sampling and Query-by-committee (QBC) are known as examples of active learning.
  • Uncertainty sampling is a method of labeling examples near the decision boundary created by the student model
  • Query-by-committee is a method of labeling examples where multiple student models give inconsistent answers. ..
  • Non-Patent Document 1 proposes a method that combines GAN (Generative Adversarial Network) and active learning.
  • GAN Geneative Adversarial Network
  • active learning In this method, GAN is used to create an artificial example in which the target classifier outputs ambiguous predictions.
  • the student model outputs an ambiguous prediction and the student model makes a mistake in the prediction.
  • the prediction of the student model may be incorrect even if it is far from the decision boundary.
  • the prediction may actually be wrong. This is especially true when the predictions of the student model are unreliable. Therefore, with the above-mentioned active learning method, it is difficult to efficiently find an example that greatly improves the prediction accuracy.
  • One object of the present invention is to efficiently find an example that greatly improves the prediction accuracy.
  • An input means that accepts training examples consisting of features
  • a label generation means for attaching a label to the training example
  • An error calculation means for generating one or more student models using the labeled training example and calculating the error between the prediction by the student model and the label.
  • An error prediction model generation means for generating an error prediction model, which is a model for predicting the error, Based on the error prediction model, an output means for outputting an example in which the error is predicted to be large, and To prepare for.
  • Another aspect of the present invention is an information processing method.
  • Accepting training examples consisting of features Label the training example and One or more student models were generated using the labeled training example, and the error between the prediction by the student model and the label was calculated.
  • An error prediction model which is a model for predicting the error, is generated. Based on the error prediction model, an example in which the error is predicted to be large is output.
  • Another aspect of the present invention is a recording medium, which is a recording medium.
  • Accepting training examples consisting of features Label the training example and One or more student models were generated using the labeled training example, and the error between the prediction by the student model and the label was calculated.
  • An error prediction model which is a model for predicting the error, is generated. Based on the error prediction model, a program that causes a computer to execute a process of outputting an example in which an error is predicted to be large is recorded.
  • a teacher model that can be regarded as outputting an absolutely correct prediction is prepared, and the prediction of the student model is evaluated by comparing it with the prediction of the teacher model.
  • the prediction of the student model is considered to be reliable.
  • the prediction of the student model is far from the prediction of the teacher model, the prediction of the student model is considered suspicious. Therefore, if an example in which the error between the prediction of the student model and the prediction of the teacher model becomes large is selected as the example for retraining, it is possible to obtain an example that greatly contributes to the improvement of accuracy.
  • FIG. 1 is a diagram conceptually showing the method of the present embodiment.
  • a teacher model that can be regarded as outputting the absolutely correct prediction described above is prepared.
  • predictions are made for each of the student model and the teacher model for multiple training examples, and the prediction error is calculated.
  • a model for estimating the prediction error between the student model and the teacher model (hereinafter referred to as "error prediction model") is generated based on the calculated error.
  • an example for retraining is selected using this error prediction model. This makes it possible to select an example that contributes to improving the accuracy of the student model.
  • FIG. 2 is a diagram conceptually showing the information processing apparatus of the present embodiment.
  • a plurality of unlabeled training examples are input to the information processing apparatus 100.
  • the information processing apparatus 100 labels the input unlabeled training example with the above-mentioned teacher model. This label corresponds to the prediction by the teacher model.
  • the information processing apparatus 100 generates a student model using the labeled training example.
  • the information processing apparatus 100 makes a prediction based on the generated student model, and generates an error prediction model showing an error between the label given by the teacher model and the prediction of the student model.
  • the error prediction model generated in this way is a model showing the prediction error between the student model and the teacher model for each example.
  • the information processing apparatus 100 outputs an example in which the error is predicted to be large by using the error prediction model.
  • FIG. 3 is a block diagram showing a hardware configuration of the information processing apparatus 100 of the first embodiment.
  • the information processing apparatus 100 includes an input IF (Interface) 11, a processor 12, a memory 13, a recording medium 14, and a database (DB) 15.
  • IF Interface
  • DB database
  • Input IF11 inputs and outputs data. Specifically, the input IF 11 acquires a training example consisting of feature quantities, and outputs an example in which an error is predicted to increase based on an error prediction model.
  • the processor 12 is a computer such as a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit), and controls the entire information processing device 100 by executing a program prepared in advance.
  • the processor 12 performs an example output process for outputting an example in which the error is predicted to be large, based on the error prediction model.
  • the memory 13 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.
  • the memory 13 stores various programs executed by the processor 12.
  • the memory 13 is also used as a working memory during execution of various processes by the processor 12.
  • the recording medium 14 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or a semiconductor memory, and is configured to be removable from the information processing device 100.
  • the recording medium 14 records various programs executed by the processor 12.
  • the DB 15 stores a training example input from the input IF 11. Further, the DB 15 stores information such as an error prediction model generated by using the training example.
  • FIG. 4 is a block diagram showing a functional configuration of the information processing apparatus 100.
  • the information processing apparatus 100 includes an input unit 21, a label generation unit 22, a prediction error calculation unit 23, an error prediction model generation unit 24, a data generation unit 25, and an output unit 26.
  • An unlabeled training example used for training a student model and a trained teacher model are input to the input unit 21.
  • the training example is composed of multidimensional features.
  • the input unit 21 outputs the unlabeled training example and the trained teacher model to the label generation unit 22. Further, the input unit 21 outputs an unlabeled training example to the prediction error calculation unit 23.
  • the label generation unit 22 uses the trained teacher model to generate a label for the input unlabeled training example, and outputs the label to the prediction error calculation unit 23. It should be noted that this label corresponds to the prediction of the teacher model for the input unlabeled training example.
  • the prediction error calculation unit 23 acquires an unlabeled training example from the input unit 21, and also acquires a label given to each training example from the label generation unit 22. As a result, the prediction error calculation unit 23 is provided with a labeled training example. The prediction error calculation unit 23 trains a student model using this labeled training example and generates a trained student model.
  • the prediction error calculation unit 23 makes a prediction using the generated student model. Then, the prediction error calculation unit 23 describes the error between the prediction by the student model and the label input from the label generation unit 22, that is, the error between the prediction by the student model and the prediction by the teacher model (hereinafter referred to as “prediction error”). Call) is calculated and output to the error prediction model generation unit 24. As will be described in detail later, in the calculation of this prediction error, an example different from the training example used for training the student model is used.
  • the error prediction model generation unit 24 acquires prediction errors for a plurality of examples from the prediction error calculation unit 23 and generates an error prediction model.
  • the error prediction model is a model that estimates the prediction error of the teacher model and the student model for each example.
  • the error prediction model generation unit 24 generates a differentiable model as an error prediction model. This is because by making the error prediction model a differentiable model, even if the student model is a non-differentiable model, it is possible to efficiently search for an example in which the error becomes large.
  • a regression model is used as the error prediction model.
  • the gradient of the regression model can be calculated to increase the prediction error. It becomes possible to discover the target.
  • FIG. 5 schematically shows a method of generating an error prediction model. It is assumed that the prediction errors E1 to E3 of the teacher model and the student model are obtained for a plurality of examples.
  • the error prediction model generation unit 24 plots each prediction error E1 to E3 in a region where the horizontal axis shows an example and the vertical axis shows a prediction error, as shown in FIG. The position of each example on the horizontal axis is determined based on the feature amount of each example. Further, it is assumed that the upper part of the vertical axis has a larger prediction error value. As shown in the figure, the prediction error of each example is discrete, and the error prediction model generation unit 24 generates a differentiable function passing through the prediction errors E1 to E3 as the error prediction model F.
  • Gaussian process regression is the observation point f (x 1 ) ,. .. .. , F (x n ) is a method of predicting a continuous function f, and more accurately, it is a method of obtaining a probability distribution of a function f from discrete observation points in the form of a Gaussian distribution.
  • FIG. 6 shows an example of the result of performing Gaussian process regression based on a plurality of observation points f (x n ) of the prediction error output by the prediction error calculation unit 23.
  • Graph f shows a true function, but it is not really known.
  • the error prediction model generation unit 24 As a result of the Gaussian process regression based on the observation point f (x n ), the average of the errors shown in the graph M and the variance of the errors shown in the gray area D are obtained. That is, the error prediction model generation unit 24 generates a graph M showing the average of the errors and a region D showing the variance of the errors as the error prediction model. In this way, the Gaussian process regression outputs a differentiable (that is, continuous) mean or variance, so that the function incorporating the mean or variance can also be differentiated, and optimization becomes easy. In a preferred example, the error prediction model generation unit 24 uses the prediction errors of the student model and the teacher model as observation points and generates an error prediction model using Gaussian process regression. The error prediction model generation unit 24 outputs the generated error prediction model to the data generation unit 25.
  • the data generation unit 25 generates an example that complements the place where the error is large based on the error prediction model, and outputs it to the output unit 26.
  • the data generation unit 25 when the function F shown in FIG. 5 is generated as an error prediction model, the data generation unit 25 generates an example P corresponding to the point p having a large error on the function F.
  • the data generation unit 25 when a training example corresponding to the example P exists among the plurality of training examples input to the input unit 21, the data generation unit 25 outputs the training example.
  • the training example corresponding to the example P does not exist, the data generation unit 25 artificially generates the feature amount corresponding to the example P based on the feature amount of the training example close to it and outputs it as the artificial example P. ..
  • the data generation unit 25 may output an example corresponding to the point where the error is maximum based on the error prediction model, or may output an example corresponding to a predetermined number of points from the one with the largest error. Often, an example corresponding to a point where the error is greater than or equal to a predetermined threshold may be output.
  • the data generation unit 25 uses the error average M as the error prediction model.
  • An example corresponding to a point where at least one of the variances D is large is output as an example in which the error is expected to be large.
  • the data generation unit 25 is an example corresponding to a predetermined number of points whose error average M or variance D is close to the maximum value, or a plurality of points whose error average M or variance D is larger than a predetermined threshold value. You may output the example corresponding to.
  • the data generation unit 25 predicts the maximum point of the true function f by Bayesian optimization using Gaussian process regression, and outputs an example corresponding to the maximum point as an example in which an error is predicted to be large. You may.
  • the output unit 26 outputs an example input from the data generation unit 25 as an example in which an error is expected to be large.
  • the example output in this way is used for retraining the student model.
  • the output example may be labeled using the teacher model used in the label generation unit 22 to be used as a training example, and may be used for training the student model.
  • the output example may be labeled by a teacher model different from the teacher model used in the label generation unit 22 or by hand.
  • a labeled training example is generated by using a teacher model for an unlabeled training example input from the input unit 21, and a student model is generated using the labeled training example. Is training. Therefore, when the prediction error calculation unit 23 calculates the prediction error of the teacher model and the student model using the same example as the training example used for training the student model, the predictions of the teacher model and the student model in each example match. Therefore, when the error prediction model generation unit 24 generates an error prediction model using this prediction error, the error prediction model has zero error at all points corresponding to the training example. It ends up. That is, overfitting occurs in the prediction error prediction itself using the error prediction model, and an error smaller than the original error is predicted. This is called "overfitting of prediction error".
  • FIG. 7 is a diagram illustrating overfitting of prediction error.
  • a plurality of points 71 indicate a training example
  • a solid line graph 72 indicates a teacher model. Since the student model is trained using the label given by the teacher model to the training example as teacher data, the prediction error with the teacher model is zero at the position of the training example 71 as shown by the graph 73 of the broken line. Be trained in. Therefore, if the prediction errors of the teacher model and the student model are calculated using the same examples as the training examples used for training the student model, the prediction errors used to generate the error prediction model are all zero, which is generalizable. It becomes impossible to generate an error prediction model.
  • the prediction error used for generating the error prediction model is calculated by using an example different from the training example used for training the student model. That is, a training example used for training a student model (hereinafter referred to as "student model training example") and an example for calculating a prediction error used for generating an error prediction model (hereinafter referred to as "error calculation example”). Is called) as a different example.
  • a training example used for training a student model hereinafter referred to as "student model training example”
  • error calculation example an example for calculating a prediction error used for generating an error prediction model
  • An error calculation example is generated by oversampling.
  • Oversampling is a method of artificially generating an example, and examples thereof include SMOTE and MUNGE.
  • all of the training examples prepared in advance are used as student model training examples to train the student model.
  • an unlabeled example x' is newly created from the training example by oversampling, and is used as an error calculation example.
  • the prediction error of the teacher model and the student model is calculated by the following formula, for example.
  • Prediction error
  • Method 2 The training example is divided and an error calculation example is generated.
  • the training example labeled by the teacher model is divided, and the divided training example is used as the student model training example to train the student model.
  • the remaining divided training examples are used as error calculation examples, and the prediction errors of the teacher model and the student model are calculated.
  • the number of data in each bootstrap sample group is N.
  • the prediction error calculation unit 23 creates a student model using each bootstrap sample group (process P3). That is, each bootstrap sample group is used as an example of student model training. As a result, M student models are generated.
  • each bootstrap sample group is generated by randomly sampling the training examples with duplication, there are samples included in the training examples but not selected in each bootstrap sample group. This is called OOB (Out-Of-Bag).
  • OOB is not included in the bootstrap sample group and is not used to generate student models. Therefore, the prediction error calculation unit 23 uses the OOB of each bootstrap sample group as an error calculation example, and calculates the prediction error of the teacher model and the student model. Specifically, the prediction error calculation unit 23 acquires predictions by the student model and the teacher model corresponding to the bootstrap sample group for the OOB of each bootstrap sample group.
  • the prediction error calculation unit 23 calculates the prediction error for each of the M bootstrap sample groups by the following formula, and outputs the average as the prediction error to the error prediction model generation unit 24 (step P4).
  • .. Prediction error
  • the prediction error between the teacher model and the student model can be calculated using an example different from the training example used to generate the student model.
  • Method 3 Obtain another training example.
  • all the training examples input to the input unit 21 are used as the student model training examples, and the student model is generated.
  • an unlabeled example different from the training example is separately acquired and used as an error calculation example.
  • FIG. 9 is a flowchart of the example output process. This process is realized by the processor 12 shown in FIG. 3 executing a program prepared in advance and operating as each element shown in FIG.
  • the input unit 21 acquires an unlabeled training example and a teacher model (step S11).
  • the label generation unit 22 assigns a label to the unlabeled training example using the teacher model (step S12).
  • the prediction error calculation unit 23 generates a student model using the training example labeled in step S12 (step S13).
  • the prediction error calculation unit 23 calculates the prediction error of the teacher model and the student model (step S14). At this time, as described above, the prediction error calculation unit 23 calculates the prediction error of the teacher model and the student model for the error calculation example different from the student model training example used for generating the student model in step S13.
  • the error prediction model generation unit 24 generates a differentialable error prediction model using the calculated prediction error by a method such as Gaussian process regression (step S15). Then, the data generation unit 25 uses the error prediction model to generate an unlabeled example in which the error is predicted to be large, and the output unit 26 outputs the generated example (step S16). Then, the process ends.
  • a method such as Gaussian process regression
  • the label generation unit 22 attaches a label to the unlabeled training example input to the input unit 21 by using a trained teacher model prepared in advance. Instead, when a labeled training example is input to the input unit 21, the label generation unit 22 may first generate a teacher model using the labeled training example. Further, the label generation unit 22 may manually assign a label instead of assigning the label using the teacher model. Further, in the above embodiment, the prediction error calculation unit 23 generates a student model using a training example to which the label generation unit 22 has a label, but instead, the trained student model prepared in advance is used. May be obtained.
  • the output unit 26 outputs an unlabeled example in which an error is expected to be large, but a labeling unit may be provided after the output unit 26.
  • the labeling unit assigns a label to the unlabeled example output by the output unit 26, so that a labeled training example that can be used for retraining the student model can be generated.
  • the label assigning unit may assign a label using the teacher model used by the label generation unit 22, and the label may be assigned using a teacher model different from the teacher model used by the label generation unit 22.
  • the label may be given, or the label may be given manually.
  • FIG. 10 is a block diagram showing a functional configuration of the information processing apparatus 50 according to the second embodiment.
  • the information processing apparatus 50 includes an input means 51, a label generation means 52, an error calculation means 53, an error prediction model generation means 54, and an output means 55.
  • the input means 51 receives a training example composed of a feature amount.
  • the label generation means 52 assigns a label to the training example.
  • the error calculation means 53 generates one or more student models using the training example to which the label is attached, and calculates the error between the prediction by the student model and the label.
  • the error prediction model generation means 54 generates an error prediction model, which is a model for predicting an error.
  • the output means 55 outputs an example in which the error is predicted to be large based on the error prediction model.
  • FIG. 11 is a flowchart of processing by the information processing apparatus 50 of the second embodiment.
  • the input means 51 receives a training example composed of a feature amount (step S21).
  • the label generation means 52 assigns a label to the training example (step S22).
  • the error calculation means 53 generates one or more student models using the training example to which the label is attached, and calculates the error between the prediction by the student model and the label (step S23).
  • the error prediction model generation means 54 generates an error prediction model, which is a model for predicting an error (step S24).
  • the output means 55 outputs an example in which the error is predicted to be large based on the error prediction model (step S25).
  • an error prediction model is generated using a training example, and an example in which an error is predicted to be large is output based on the error prediction model. Therefore, the accuracy of the student model can be efficiently improved by retraining the student model using the output example.
  • An input means that accepts training examples consisting of features
  • a label generation means for attaching a label to the training example
  • An error calculation means for generating one or more student models using the labeled training example and calculating the error between the prediction by the student model and the label.
  • An error prediction model generation means for generating an error prediction model, which is a model for predicting the error, Based on the error prediction model, an output means for outputting an example in which the error is predicted to be large, and Information processing device equipped with.
  • Appendix 2 The information processing apparatus according to Appendix 1, wherein the error prediction model generation means generates a differentiable error prediction model based on the error between the prediction of the student model and the label for the plurality of training examples.
  • the error prediction model is a regression model and The information processing apparatus according to Appendix 1 or 2, wherein the output means predicts an error in the above example based on a gradient in the regression model.
  • the error prediction model is a model that outputs the differentiable mean and variance of the error.
  • the label generation means assigns a label to the training example by using the teacher model generated by using the training example.
  • the information processing apparatus according to any one of Supplementary note 1 to 4, wherein the error calculation means calculates an error between the label, which is a prediction by the teacher model, and the prediction by the student model.
  • the error calculating means generates the one or more student models by using at least a part of the examples of the training example, and calculates the error by using an example different from the example used to generate the student model.
  • the information processing apparatus according to any one of Supplementary note 1 to 5.
  • the error calculation means generates a plurality of sample groups by random sampling with duplication from the training example, generates a student model using each of the sample groups, and is included in the training example for each student model.
  • the information processing apparatus according to any one of Supplementary note 1 to 5, wherein the error is calculated for the data not included in the sample group, and the average of the errors calculated for the plurality of student models is calculated.
  • Appendix 8 Accepting training examples consisting of features Label the training example and One or more student models were generated using the labeled training example, and the error between the prediction by the student model and the label was calculated.
  • An error prediction model which is a model for predicting the error, is generated.
  • An information processing method that outputs an example in which an error is predicted to increase based on the error prediction model.
  • An error prediction model which is a model for predicting the error, is generated.
  • a recording medium recording a program that causes a computer to execute a process of outputting an example in which an error is predicted to be large based on the error prediction model.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Algebra (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
PCT/JP2020/044483 2020-11-30 2020-11-30 情報処理装置、情報処理方法、及び、記録媒体 Ceased WO2022113338A1 (ja)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP2022564995A JP7529040B2 (ja) 2020-11-30 2020-11-30 情報処理装置、情報処理方法、及び、プログラム
PCT/JP2020/044483 WO2022113338A1 (ja) 2020-11-30 2020-11-30 情報処理装置、情報処理方法、及び、記録媒体
US18/037,298 US20240020575A1 (en) 2020-11-30 2020-11-30 Information processing device, information processing method, and recording medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2020/044483 WO2022113338A1 (ja) 2020-11-30 2020-11-30 情報処理装置、情報処理方法、及び、記録媒体

Publications (1)

Publication Number Publication Date
WO2022113338A1 true WO2022113338A1 (ja) 2022-06-02

Family

ID=81755505

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/044483 Ceased WO2022113338A1 (ja) 2020-11-30 2020-11-30 情報処理装置、情報処理方法、及び、記録媒体

Country Status (3)

Country Link
US (1) US20240020575A1 (https=)
JP (1) JP7529040B2 (https=)
WO (1) WO2022113338A1 (https=)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025013478A1 (ja) * 2023-07-10 2025-01-16 パナソニックＩｐマネジメント株式会社 評価装置、評価方法、およびプログラム

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016528549A (ja) * 2013-08-07 2016-09-15 エーエスエムエル ネザーランズ ビー．ブイ． メトロロジ方法および装置、リソグラフィシステムならびにデバイス製造方法
JP2019159654A (ja) * 2018-03-12 2019-09-19 国立研究開発法人情報通信研究機構 時系列情報の学習システム、方法およびニューラルネットワークモデル
WO2020162048A1 (ja) * 2019-02-07 2020-08-13 国立大学法人山梨大学 信号変換システム、機械学習システムおよび信号変換プログラム

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210117828A1 (en) * 2018-06-27 2021-04-22 Sony Corporation Information processing apparatus, information processing method, and program
JP7143148B2 (ja) * 2018-08-23 2022-09-28 三菱重工業株式会社 予測装置、予測方法、及びプログラム

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2016528549A (ja) * 2013-08-07 2016-09-15 エーエスエムエル ネザーランズ ビー．ブイ． メトロロジ方法および装置、リソグラフィシステムならびにデバイス製造方法
JP2019159654A (ja) * 2018-03-12 2019-09-19 国立研究開発法人情報通信研究機構 時系列情報の学習システム、方法およびニューラルネットワークモデル
WO2020162048A1 (ja) * 2019-02-07 2020-08-13 国立大学法人山梨大学 信号変換システム、機械学習システムおよび信号変換プログラム

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
INATSU, YU ET AL.: "Active learning for identifying local minimum points based on the derivative of Gaussian process", IEICE TECHNICAL REPORT, vol. 118, no. 284, 5 November 2018 (2018-11-05), pages 373 - 380 *
ISHIGAMI, S.: "RandomForest R package", 1 February 2018 (2018-02-01), Retrieved from the Internet <URL:https://www.slideshare.net/ShumaIshigami/randomforestr-package/34> [retrieved on 20210121] *
YOO DONGGEUN; KWEON IN SO: "Learning Loss for Active Learning", 2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), IEEE, 15 June 2019 (2019-06-15), pages 93 - 102, XP033687105, DOI: 10.1109/CVPR.2019.00018 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2025013478A1 (ja) * 2023-07-10 2025-01-16 パナソニックＩｐマネジメント株式会社 評価装置、評価方法、およびプログラム

Also Published As

Publication number Publication date
JP7529040B2 (ja) 2024-08-06
US20240020575A1 (en) 2024-01-18
JPWO2022113338A1 (https=) 2022-06-02

Similar Documents

Publication Publication Date Title
US12106223B2 (en) Data valuation using reinforcement learning
US12205348B2 (en) Neural network orchestration
US12585979B2 (en) Managing data drift and outliers for machine learning models trained for image classification
WO2021000958A1 (zh) 用于实现模型训练的方法及装置、计算机存储介质
JP6525002B2 (ja) メンテナンス時期決定装置、劣化予測システム、劣化予測方法および記録媒体
CN105046045B (zh) 一种基于贝叶斯组合的Web服务QoS预测方法
CN110750455B (zh) 基于系统日志分析的智能在线自更新故障诊断方法和系统
Hammami et al. On-line self-adaptive framework for tailoring a neural-agent learning model addressing dynamic real-time scheduling problems
JP2023527237A (ja) 産業プロセスモデル生成システム
CN117155806A (zh) 一种通信基站流量预测方法与装置
WO2022113338A1 (ja) 情報処理装置、情報処理方法、及び、記録媒体
JP7821028B2 (ja) 推論装置、生成方法、および生成プログラム
US20250054584A1 (en) Computer-readable recording medium storing active learning program, method, and apparatus
CN119129768B (zh) 一种面向非完整多元时间序列预测的学习方法
JP7439923B2 (ja) 学習方法、学習装置及びプログラム
CN120374331A (zh) 一种基于地理信息的数字孪生智慧校园管理方法及系统
JP7529041B2 (ja) 情報処理装置、情報処理方法、及び、プログラム
WO2024180789A1 (ja) 情報処理装置、情報処理方法、プログラム
US20230196088A1 (en) Fan behavior anomaly detection using neural network
CN114418118A (zh) 决策能力约束下多源网络流数据预测方法和装置
Gromov et al. Multi-Step-Ahead Prediction of Chaotic Time Series: Self-Healing Algorithm for Restoring Values at Non-Predictable Points
JP7475549B2 (ja) 学習装置、予測装置、予測システム、学習方法、予測方法、及び予測プログラム
CN113077040A (zh) 时间序列预测方法及相关的时间序列预测电路
CN119988120A (zh) 一种基于迁移学习的系统性能自动化测试与优化方法
JP2026003425A (ja) 情報処理装置、情報処理方法、及びプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20963602

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18037298

Country of ref document: US

ENP Entry into the national phase

Ref document number: 2022564995

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20963602

Country of ref document: EP

Kind code of ref document: A1