公开(公告)号：US20220292345A1

公开(公告)日：2022-09-15

申请号：US17392261

申请日：2021-08-03

Applicant: NEC CORPORATION

Inventor： Vivek BARSOPIA ,  Yoshio KAMEDA ,  Tomoya SAKAI

IPC: G06N3/08 ,  G06N3/04

Abstract: Distributionally robust models are obtained by operations including training, according to a loss function, a first learning function with a training data set to produce a first model, the training data set including a plurality of samples. The operations may further include training a second learning function with the training data set to produce a second model, the second model having a higher accuracy than the first model. The operations may further include assigning an adversarial weight to each sample among the plurality of samples set based on a difference in loss between the first model and the second model. The operations may further include retraining, according to the loss function, the first learning function with the training data set to produce a distrtibutionally robust model, wherein during retraining the loss function further modifies loss associated with each sample among the plurality of samples based on the assigned adversarial weight.

公开(公告)号：US20220027760A1

公开(公告)日：2022-01-27

申请号：US17297181

申请日：2018-12-10

Applicant: NEC Corporation

Inventor： Tomoya SAKAI ,  Yasuhiro SOGAWA

IPC: G06N5/04 ,  G06K9/62

Abstract: A learning device 100 includes correspondence inference unit which calculates outputs of predictors, which have learned for seen tasks or seen classes, for test input data, and infers correspondences between the calculated outputs and attribute information corresponding to an unseen task or an unseen class, and prediction unit which calculates a prediction output for the attribute information corresponding to the unseen task or the unseen class, using the inferred correspondences.

公开(公告)号：US20240038389A1

公开(公告)日：2024-02-01

申请号：US18037510

申请日：2020-11-25

Applicant: NEC Corporation

Inventor： Tomoya SAKAI ,  Hiroshi TAMANO

IPC: G16H50/20

CPC classification number: G16H50/20

Abstract: In the strange feeling prediction device, the prediction result acquisition means inputs diagnosis data to a target prediction model which is a trained prediction model serving as a target, and acquires a prediction result by the target prediction model. The label acquisition means acquires a strange feeling label indicating a strange feeling of an expert with respect to the prediction result. The strange feeling prediction model training means trains a strange feeling prediction model using the prediction result and the strange feeling label. The strange feeling prediction means outputs a strange feeling index indicating the strange feeling with respect to the prediction result outputted by the target prediction model, using the trained strange feeling prediction model.

公开(公告)号：US20220343212A1

公开(公告)日：2022-10-27

申请号：US17389237

申请日：2021-07-29

Applicant: NEC CORPORATION

Inventor： Vivek BARSOPIA ,  Yoshio KAMEDA ,  Tomoya SAKAI

IPC: G06N20/00

Abstract: Forward compatible models are obtained by operations including training a learning function with a current training data set to produce a first model, the current training data set including a plurality of samples, generating a plurality of prospective models, each prospective model based on a variation of one of the current training data set or the first model, adjusting a plurality of sample weights based on output of one or more prospective models among the plurality of prospective models in response to input of the current training data set, and retraining the learning function with the current training data set and the plurality of sample weights to produce a second model.

公开(公告)号：US20240119357A1

公开(公告)日：2024-04-11

申请号：US18276809

申请日：2021-02-25

Applicant: NEC Corporation

Inventor： Keita SAKUMA ,  Tomoya SAKAI ,  Yoshio KAMEDA ,  Hiroshi TAMANO

IPC: G06N20/00

CPC classification number: G06N20/00

Abstract: Provided are an analysis device, an analysis method, and a program capable of easily identifying a factor of a prediction error in prediction using a prediction model on the basis of various viewpoints. An analysis device (1) includes: a metric evaluation unit (2) that calculates and evaluates a plurality of types of metrics with respect to a prediction model, data of explanatory variables used in the prediction model, or data of target variables used in the prediction model; and a factor identification unit (3) that identifies a factor of an error in prediction by the prediction model according to a combination of evaluation results of the plurality of types of metrics.

公开(公告)号：US20230195851A1

公开(公告)日：2023-06-22

申请号：US17925880

申请日：2020-05-28

Applicant: NEC Corporation

Inventor： Genki KUSANO ,  Tomoya SAKAI

IPC: G06F18/2431

CPC classification number: G06F18/2431

Abstract: A data classification system calculates, for each known class, which is a class indicated in training data used for performing learning of class classification, a known class likelihood indicating a likelihood of target data belonging to the known class among all known classes. The data classification system selects, as candidates for a class to which the target data belongs, classes in which at least one of the known classes is excluded based on the known class likelihood among all classes in the class classification. The data classification system calculates, at least for each of the classes included in the candidates, an all-class likelihood indicating a likelihood of the target data belonging to the class among all the classes. The data classification system estimates the class to which the target data belongs as any one of the classes among the candidates, based on the all-class likelihood.

公开(公告)号：US20220092622A1

公开(公告)日：2022-03-24

申请号：US17420026

申请日：2019-01-10

Applicant: NEC Corporation

Inventor： Tomoya SAKAI ,  Yasuhiro SOGAWA

IPC: G06Q30/02 ,  G06N5/04

Abstract: The learning unit 81 learns an attribute viewpoint model in which attributes of a target are explanatory variables for each target person so as to minimize a difference between a prediction result by a predictor that predicts an evaluation result of each target person based on a feature vector of the target person and a prediction result by a prediction model that predicts an evaluation result learned for each target person, using the attributes of the target as explanatory variables. The attribute generation unit 82 generates an attribute so that an evaluation result obtained according to the attribute applied to the learned prediction model satisfies the specified objective.

公开(公告)号：US20240119296A1

公开(公告)日：2024-04-11

申请号：US18276290

申请日：2021-06-07

Applicant: NEC Corporation

Inventor： Akira TANIMOTO ,  Tomoya SAKAI ,  Takashi TAKENOUCHI ,  Hisashi KASHIMA

IPC: G06N3/09

CPC classification number: G06N3/09

Abstract: A learning device calculates an estimation target item reference value according to a fixed value of each estimation target object. The learning device acquires learning data that includes the fixed value of each estimation target object, a variable item value, and an estimation target item value according to the fixed value and the variable item value. The learning device trains, using the learning data and an evaluation function, a model that outputs an estimated value of the estimation target item value in response to input of the fixed value of each estimation target object and the variable item value.

公开(公告)号：US20240028912A1

公开(公告)日：2024-01-25

申请号：US17863338

申请日：2022-07-12

Applicant: NEC CORPORATION

Inventor： Vivek BARSOPIA ,  Yoshio KAMEDA ,  Tomoya SAKAI ,  Keita SAKUMA ,  Ryuta MATSUNO

IPC: G06N5/02

CPC classification number: G06N5/022

Abstract: Predictively robust models are trained by embedding a distribution of each temporal data set among a plurality of temporal data sets into a feature vector, predicting a future feature vector of a distribution of a future data set, based on the feature vector of each temporal data set among a plurality of temporal data sets, creating the future data set from the future feature vector, perturbing the future data set to produce a plurality of perturbed future data sets, and training a learning function using the future data set and each perturbed future data set to produce a model.

公开(公告)号：US20220092475A1

公开(公告)日：2022-03-24

申请号：US17419974

申请日：2019-01-11

Applicant: NEC Corporation

Inventor： Yasuhiro SOGAWA ,  Tomoya SAKAI

IPC: G06N20/00 ,  G06K9/62

Abstract: A target task attribute estimation unit 81 estimates an attribute vector of an existing predictor based on samples in a domain of a target task, and estimates an attribute vector of the target task based on a transformation method for transforming labeled samples into a space consisting of the estimated attribute vector based on a result of applying the labeled samples of the target task to the predictor. A prediction value calculation unit 82 calculates a prediction value of a prediction target sample to be transformed by the transformation method based on the attribute vector of the target task.