RTFN: A Robust Temporal Feature Network for Time Series Classification

Abstract

Time series data usually contains local and global patterns. Most of the existing feature networks pay more attention to local features rather than the relationships among them. The latter is, however, also important yet more difficult to explore. To obtain sufficient representations by a feature network is still challenging. To this end, we propose a novel robust temporal feature network (RTFN) for feature extraction in time series classification, containing a temporal feature network (TFN) and an LSTM-based attention network (LSTMaN). TFN is a residual structure with multiple convolutional layers. It functions as a local-feature extraction network to mine sufficient local features from data. LSTMaN is composed of two identical layers, where attention and long short-term memory (LSTM) networks are hybridized. This network acts as a relation extraction network to discover the intrinsic relationships among the extracted features at different positions in sequential data. In experiments, we embed RTFN into a supervised structure as a feature extractor and into an unsupervised structure as an encoder, respectively. The results show that the RTFN-based structures achieve excellent supervised and unsupervised performance on a large number of UCR2018 and UEA2018 datasets.

41pages, 7figures, Revised Paper

References in corpus (14)

• Natural Scales in Geographical Patterns
• YOLOv3: An Incremental Improvement
• An overview of gradient descent optimization algorithms
• Deep learning for time series classification: a review
• Multivariate LSTM-FCNs for Time Series Classification
• Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks
• Insights into LSTM Fully Convolutional Networks for Time Series Classification
• Combining time-series and textual data for taxi demand prediction in event areas: a deep learning approach
• Adversarial Attacks on Deep Neural Networks for Time Series Classification
• Multivariate Time Series Classification with WEASEL+MUSE
• Deep Temporal Clustering : Fully Unsupervised Learning of Time-Domain Features
• The Heterogeneous Ensembles of Standard Classification Algorithms (HESCA): the Whole is Greater than the Sum of its Parts
• Ultra-Fast Shapelets for Time Series Classification
• Residual Attention Net for Superior Cross-Domain Time Sequence Modeling