公开(公告)号：WO2022216792A1

公开(公告)日：2022-10-13

申请号：PCT/US2022/023620

申请日：2022-04-06

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： MIZOGUCHI, Takehiko ,  LUMEZANU, Cristian ,  CHEN, Yuncong ,  CHEN, Haifeng

IPC: G06N3/04 ,  G06F11/30 ,  G06N3/08

Abstract: Methods and systems for training a model include training a feature extraction model to extract a feature vector from a multivariate time series segment, based on a set of training data corresponding to measurements of a system in a first domain. Adapting the feature extraction model to a second domain, based on prototypes of the training data in the first domain and new time series data corresponding to measurements of the system in a second domain.

公开(公告)号：WO2021015936A1

公开(公告)日：2021-01-28

申请号：PCT/US2020/040649

申请日：2020-07-02

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： CHEN, Yuncong ,  YUAN, Hao ,  SONG, Dongjin ,  LUMEZANU, Cristian ,  CHEN, Haifeng ,  MIZOGUCHI, Takehiko

IPC: G06F16/33 ,  G06F16/35 ,  G06N3/08

Abstract: A system (200) for cross-modal data retrieval is provided that includes a neural network having a time series encoder (211) and text encoder (212) which are jointly trained using an unsupervised training method which is based on a loss function. The loss function jointly evaluates a similarity of feature vectors of training sets of two different modalities of time series and free-form text comments and a compatibility of the time series and the free-form text comments with a word-overlap-based spectral clustering method configured to compute pseudo labels for the unsupervised training method. The computer processing system further includes a database (205) for storing the training sets with feature vectors extracted from encodings of the training sets. The encodings are obtained by encoding a training set of the time series using the time series encoder and encoding a training set of the free-form text comments using the text encoder.

公开(公告)号：WO2022216599A1

公开(公告)日：2022-10-13

申请号：PCT/US2022/023299

申请日：2022-04-04

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： CHEN, Yuncong ,  LUMEZANU, Cristian ,  CHENG, Wei ,  MIZOGUCHI, Takehiko ,  NATSUMEDA, Masanao ,  CHEN, Haifeng

IPC: G06F11/30 ,  G06F40/284 ,  G06F40/169 ,  G06N20/00 ,  G06N3/04

Abstract: A method for explaining sensor time series data in natural language is presented. The method includes training (1001) a neural network model with text-annotated time series data, the neural network model including a time series encoder and a text generator, allowing (1003) a human operator to select a time series segment from the text-annotated time series data, the time series segment processed by the time series encoder, outputting (1005), from the time series encoder, a sequence of hidden state vectors, one for each timestep, and generating (1007) readable explanatory texts for the human operator based on the selected time series segment, the readable explanatory texts being a set of comment texts explaining and interpreting the selected time series segment in a plurality of different ways.

公开(公告)号：WO2022032090A1

公开(公告)日：2022-02-10

申请号：PCT/US2021/044933

申请日：2021-08-06

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： SONG, Dongjin ,  CHEN, Yuncong ,  LUMEZANU, Cristian ,  MIZOGUCHI, Takehiko ,  CHEN, Haifeng ,  ZHU, Wei

IPC: G06N20/20 ,  G06N3/04 ,  G06N3/08

Abstract: Methods and systems for training a neural network include collecting (302) model exemplar information from edge devices, each model exemplar having been trained using information local to the respective edge devices. The collected model exemplar information is aggregated (304) together using federated averaging. Global model exemplars are trained (306) using federated constrained clustering. The trained global exemplars are transmitted (206) to respective edge devices.

公开(公告)号：WO2021011205A1

公开(公告)日：2021-01-21

申请号：PCT/US2020/040629

申请日：2020-07-02

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： CHEN, Yuncong ,  SONG, Dongjin ,  LUMEZANU, Cristian ,  CHEN, Haifeng ,  MIZOGUCHI, Takehiko

IPC: G06F16/33 ,  G06F16/332 ,  G06F16/36 ,  G06N3/08 ,  G06N3/063

Abstract: A system (200) for cross-modal data retrieval is provided which includes a neural network having a time series encoder (211) and text encoder jointly trained based on a triplet loss relating to two different modalities of (i) time series and (ii) free-form text comments. A database (205) stores training sets with feature vectors extracted from encodings of the training sets. The encodings are obtained by encoding the time series using the time series encoder and encoding the text comments using the text encoder. A processor retrieves the feature vectors corresponding to at least one of the modalities from the database for insertion into a feature space together with a feature vector corresponding to a testing input relating to at least one of a testing time series and a testing free-form text comment, determines a set of nearest neighbors from among the feature vectors based on distance criteria, and outputs testing results.

公开(公告)号：WO2022216375A1

公开(公告)日：2022-10-13

申请号：PCT/US2022/017429

申请日：2022-02-23

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： NATSUMEDA, Masanao ,  CHENG, Wei ,  MIZOGUCHI, Takehiko ,  CHEN, Haifeng

IPC: G06N3/04 ,  G06N3/08

Abstract: Methods and systems for training a neural network include training models for respective sensor groups in a cyber-physical system. Combinations of sensor groups and operational modes are sampled. A combination model is trained for each of the sampled combinations. A best combination model is determined based on performance measured during training. The best combination model is fine-tuned.

公开(公告)号：WO2022010731A1

公开(公告)日：2022-01-13

申请号：PCT/US2021/040081

申请日：2021-07-01

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： MIZOGUCHI, Takehiko ,  SONG, Dongjin ,  CHEN, Yuncong ,  LUMEZANU, Cristian ,  CHEN, Haifeng

IPC: G06F16/28 ,  G06F16/25 ,  G06F16/22 ,  G06F16/2458 ,  G06N3/08 ,  G06N3/04 ,  G06N20/00 ,  G06K9/0053 ,  G06K9/6215 ,  G06K9/6232 ,  G06K9/6255 ,  G06K9/6261 ,  G06K9/6277 ,  G06N3/0445

Abstract: Systems and methods for retrieving similar multivariate time series segments are provided. The systems and methods include extracting (920) a long feature vector and a short feature vector from a time series segment, converting (930) the long feature vector into a long binary code, and converting (930) the short feature vector into a short binary code. The systems and methods further include obtaining (940) a subset of long binary codes from a binary dictionary storing dictionary long codes based on the short binary codes, and calculating (950) similarity measure for each pair of the long feature vector with each dictionary long code. The systems and methods further include identifying (960) a predetermined number of dictionary long codes having the similarity measures indicting a closest relationship between the long binary codes and dictionary long codes, and retrieving (970) a predetermined number of time series segments associated with the predetermined number of dictionary long codes.

公开(公告)号：WO2021207369A1

公开(公告)日：2021-10-14

申请号：PCT/US2021/026196

申请日：2021-04-07

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： SONG, Dongjin ,  MIZOGUCHI, Takehiko ,  LUMEZANU, Cristian ,  CHEN, Haifeng

IPC: G05B23/02 ,  G06Q50/04

Abstract: Methods and systems for detecting and correcting anomalies includes generating (206) historical binary codes from historical time series segments. The historical time series segments are each made up of measurements from respective sensors. A latest binary code is generated (208) from a latest time series segment. It is determined that the latest time series segment represents anomalous behavior, based on a comparison of the latest binary code to the historical binary codes. The sensors are ranked (214), based on a comparison of time series data of the sensors in the latest time series segment to respective time series data of the historical time series, to generate (216) a sensor ranking. A corrective action is performed (218) responsive to the detected anomaly, prioritized according to the sensor ranking.

公开(公告)号：WO2021041631A1

公开(公告)日：2021-03-04

申请号：PCT/US2020/048139

申请日：2020-08-27

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： SONG, Dongjin ,  CHEN, Yuncong ,  LUMEZANU, Cristian ,  MIZOGUCHI, Takehiko ,  CHEN, Haifeng ,  ZHU, Dixian

IPC: G06N3/08 ,  G06N3/04 ,  G06T9/00 ,  G06F11/30

Abstract: A computer-implemented method for monitoring computing system status by implementing a deep unsupervised binary coding network includes receiving (41) multivariate time series data from one or more sensors associated with a system, implementing (420) a long short-term memory (LSTM) encoder-decoder framework to capture temporal information of different time steps within the multivariate time series data and perform binary coding, the LSTM encoder-decoder framework including a temporal encoding mechanism, a clustering loss and an adversarial loss, computing (430) a minimal distance from the binary code to historical data, and obtaining (440) a status determination of the system based on a similar pattern analysis using the minimal distance.

公开(公告)号：WO2022072772A1

公开(公告)日：2022-04-07

申请号：PCT/US2021/053078

申请日：2021-10-01

Applicant: NEC LABORATORIES AMERICA, INC.

Inventor： CHEN, Yuncong ,  CHEN, Zhengzhang ,  LUMEZANU, Cristian ,  NATSUMEDA, Masanao ,  YU, Xiao ,  CHENG, Wei ,  MIZOGUCHI, Takehiko ,  CHEN, Haifeng

IPC: G06N3/08 ,  G06N3/04 ,  G06N5/02

Abstract: A method for system metric prediction and influential events identification by concurrently employing metric logs and event logs is presented. The method includes concurrently modeling (1301) multivariate metric series and individual events in event series by a multi-stream recurrent neural network (RNN) to improve prediction of future metrics, where the multi- stream RNN includes a series of RNNs, one RNN for each metric and one RNN for each event sequence and modeling (1303) causality relations between the multivariate metric series and the individual events in the event series by employing an attention mechanism to identify target events most responsible for fluctuations of one or more target metrics.