Modeling Time Series Similarity with Siamese Recurrent Networks

Abstract

Traditional techniques for measuring similarities between time series are based on handcrafted similarity measures, whereas more recent learning-based approaches cannot exploit external supervision. We combine ideas from time-series modeling and metric learning, and study siamese recurrent networks (SRNs) that minimize a classification loss to learn a good similarity measure between time series. Specifically, our approach learns a vectorial representation for each time series in such a way that similar time series are modeled by similar representations, and dissimilar time series by dissimilar representations. Because it is a similarity prediction models, SRNs are particularly well-suited to challenging scenarios such as signature recognition, in which each person is a separate class and very few examples per class are available. We demonstrate the potential merits of SRNs in within-domain and out-of-domain classification experiments and in one-shot learning experiments on tasks such as signature, voice, and sign language recognition.

arXiv comment11 pages

References in corpus (4)

• Neural Machine Translation by Jointly Learning to Align and Translate
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Results
• Sign Language Tutoring Tool

Cited by in corpus (7)

• Deep Recurrent Convolutional Networks for Video-based Person Re-identification: An End-to-End Approach
• NeuralWarp: Time-Series Similarity with Warping Networks
• Autowarp: Learning a Warping Distance from Unlabeled Time Series Using Sequence Autoencoders
• Human Motion Analysis with Deep Metric Learning
• OSVNet: Convolutional Siamese Network for Writer Independent Online Signature Verification
• Learning Correlation Space for Time Series
• Dynamic State Warping