Abstract: A method of learning a neural network, wherein the neural network includes: a feature selection layer for selecting a part of input data including information about a domain of each sample; a feature extraction layer for extracting a feature quantity on the basis of the selected input data; and a prediction layer for performing a prediction on the basis of the feature quantity, and the method includes adjusting a weight parameter of the neural network to increase a prediction accuracy by the prediction layer and to reduce a contribution to a prediction result of the prediction layer by the domain of the input data.

Abstract: A method of learning a neural network, wherein the neural network includes: a feature selection layer for selecting a part of input data; a feature extraction layer for extracting a feature quantity on the basis of the selected input data; a prediction layer for performing a prediction on the basis of the feature quantity; and a partial reconstruction layer for reconstructing the selected input data on the basis of the feature quantity, and the method includes adjusting a weight parameter of the neural network on the basis of a prediction accuracy by the prediction layer and a reconstruction error in the partial reconstruction layer.

Abstract: To extract data-for-machine-leaning so as to enable accurate detection of an abnormality while reducing normal state patterns in a mobile network. A learning data extracting apparatus includes: an issuing section that issues a network service use request which is to be processed by cooperation of a plurality of communication apparatuses constituting C-plane; an obtaining section that obtains, as pieces of candidate learning data, time series data of information relating to the plurality of communication apparatuses; and an extracting section that extracts, as data-for-machine-leaning, a piece of candidate learning data with which the process in response to the network service use request has been successfully ended, from among the pieces of candidate learning data.

Abstract: Provided is a method of learning a neural network that predicts a remaining life of a target device that is a maintenance target. The neural network includes: (i) a first model for predicting a remaining life at an arbitrary time of maintenance cycle data, as a value based on an arbitrary reference value; and (ii) a second model for predicting a remaining life at a final time of the maintenance cycle data, as a value based on the reference value. The method comprises updating a weight parameter of the first model so as to predict a remaining life based on an end of the maintenance cycle data, by using an output of the first model and an output of the second model that are obtained from learning data including a plurality of maintenance cycle data.

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 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 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.

Abstract: Provided is an analysis system including: an analysis unit including a classifier that performs classification of an event type on input time-series data; a display information generation unit that generates first display information used for displaying, out of the time-series data, first time-series data in which association of an event type is undecided and which is classified by the classifier as a first event type corresponding to a state where a target event is occurring, second time-series data associated with the first event type, and third time-series data associated with a second event type corresponding to a state where the target event is not occurring; and an input unit that accepts first input regarding association of an event type with the first time-series data.

Abstract: A failure prediction system is provided. The system includes a model-based signature generator generating feature vectors from individual attributes of multi-variate time series data based on sequence importance and attribute importance. The system further includes a knowledge database storing feature vectors corresponding to a set of different failure types. The system also includes a set of similarity detectors. Each detect any of the feature vectors generated by the model-based signature generator that are similar to any of the feature vectors corresponding to a respective one of the different failure types stored in the knowledge database based on a similarity threshold and output the respective one of the different failure types and a likely time period when the respective one of the different failure types will occur.

Abstract: A learning system including: a training data acquisition unit that, for each of a plurality of feature amounts obtained by converting time-series data on a predetermined period basis, acquires training data with which an event type in the predetermined period is associated; and a classifier learning unit that performs learning on a plurality of classifiers different from each other by using a feature amount group including one or more feature amounts corresponding to some or all periods out of the plurality of feature amounts included in the training data to perform learning on a classifier, which classifies the event type, for multiple times while changing a combination in the feature amount group, and the event type includes a first event type corresponding to a state where a target event to be classified occurs and a second event type corresponding to a state where the target event does not occur.

Abstract: Methods and systems for predicting failure in a cyber-physical system include determining a prediction index based on a comparison of input time series, from respective sensors in a cyber-physical system, to failure precursors. A failure precursor is detected in the input time series, responsive to a comparison of the prediction index to a threshold. A subset of the sensors associated with the failure precursor is determined, based on a gradient of the prediction index. A corrective action is performed responsive to the determined subset of sensors.

Abstract: Methods and systems for preventing spacecraft damage include identifying a space weather event that corresponds to a spacecraft system failure. A spacecraft system is determined that causes the spacecraft system failure, triggered by the space weather event. A corrective action is performed on the determined spacecraft system to prevent spacecraft system failures from being triggered by future space weather events.

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.

Abstract: A method for explaining sensor time series data in natural language is presented. The method includes training a neural network model with text-annotated time series data, the neural network model including a time series encoder and a text generator, allowing 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, from the time series encoder, a sequence of hidden state vectors, one for each timestep, and generating 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.

Abstract: A system analysis method includes: acquiring history information indicating, based on sensor values outputted by sensors, whether one of sensor values outputted by respective sensors indicates abnormality and/or whether individual relationship between sensor values outputted by different sensors indicates abnormality in time-series manner; estimating a change point group of change points, each indicating a time point system state has changed, based on history information; estimating relevance levels, each indicating relevance to the system state between two arbitrary time points included in the change point group; generating groups of change point groups by classifying the change point group into a plurality of groups based on the history information and the relevance levels; and generating and outputting output information, as information relating abnormality per group of the change point groups.

Abstract: Systems and methods for fault detection in a sensor network include receiving sensor data from sensors in the sensor network with a communication device. The sensor data is analyze to determine if the sensor data is indicative of a fault with a fault detection model, the fault detection model including; predicting the sensor data with an autoencoder by encoding the sensor data and decoding encoded the sensor data, autoregressively model the sensor data with an autoregressor, combining the modeled sensor data and the predicted sensor data with a combiner to produce reconstructed sensor data, and comparing the reconstructed sensor data to the sensor data with an anomaly evaluator to determine anomalies. An anomaly classification is produced by comparing the anomalies to historical anomalies with an anomaly classifier. Faults in the sensor network are automatically mitigated with a processing device based on the anomaly classification.

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 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 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.

Abstract: Systems and methods for detecting anomalies in a plurality of showcases are provided. A system can obtain a corresponding table between each of the plurality of showcases and at least one corresponding sensor. The system obtains information for showcase clustering. The system can include a processor device that can determine at least one cluster of showcases based on the information for showcase clustering and the corresponding table between each of the plurality of showcases and the at least one corresponding sensor. The system can build at least one model for each of the at least one cluster of showcases and detect at least one anomaly based on data from the at least one cluster of showcases and the at least one model.

Abstract: Systems and methods for determining a remaining useful life of a system. The system and method include one or more processors; a memory coupled to the one or more processors; a data acquisition unit configured to receive run-to-failure time series data; a neural network training unit configured to train a neural network model to determine a point in time that a health index changes from a healthy stage to a degradation stage; a remaining useful life estimation unit configured to estimate a first remaining useful life of the system based on the point in time; estimate a second remaining useful life of the system by converting a feature representation output by the second neural network; minimize the difference between the first remaining useful life and the second remaining useful life; classify the health stage based on a probability; and an output unit configured to send a warning to a user.

Abstract: Systems and methods for early anomaly prediction on multi-variate time series data are provided. The method includes identifying a user labeled abnormal time period that includes at least one anomaly event. The method also includes determining a multi-variate time series segment of multivariate time series data that occurs before the user labeled abnormal time period, and treating, by a processor device, the multi-variate time series segment to include precursor symptoms of the at least one anomaly event. The method includes determining instance sections from the multi-variate time series segment and determining at least one precursor feature vector associated with the at least one anomaly event for at least one of the instance sections based on applying long short-term memory (LSTM). The method further includes dispatching predictive maintenance based on the at least one precursor feature vector.

Abstract: Provided is an analysis system including: an analysis unit including a classifier that performs classification of an event type on input time-series data; a display information generation unit that generates first display information used for displaying, out of the time-series data, first time-series data in which association of an event type is undecided and which is classified by the classifier as a first event type corresponding to a state where a target event is occurring, second time-series data associated with the first event type, and third time-series data associated with a second event type corresponding to a state where the target event is not occurring; and an input unit that accepts first input regarding association of an event type with the first time-series data.

Abstract: A computer-implemented method is provided for hardware management based on estimating a Remaining Useful Life (RUL) of an object. The method includes estimating the RUL at a time point preceding the RUL by two RUL estimation methods respectively applied to a first time series subsequence and a second time series subsequence from among an overall time series sequence. The first time series subsequence includes run-to-event data used in a first one of the two RUL estimation methods. The second time series subsequence is applied to a leaky truncated RUL function as a second one of the two RUL estimation methods to obtain a model for RUL estimation. The method further includes estimating the RULE at inference using the first one of the two RUL estimation methods. The method also includes selectively servicing or replacing the object with a replacement object responsive to the RUL.