Deep Attentive Time Warping

TL;DR

This paper introduces a neural network model with an attention mechanism for adaptive time warping in time series, improving robustness and discriminative power over traditional DTW, especially in online signature verification.

Contribution

It presents a novel learnable time warping model using bipartite attention, trained with metric learning and pre-trained with DTW for enhanced performance.

Findings

Outperforms DTW in accuracy for time series similarity

Achieves state-of-the-art results in online signature verification

Demonstrates robustness to nonlinear time distortions

Abstract

Similarity measures for time series are important problems for time series classification. To handle the nonlinear time distortions, Dynamic Time Warping (DTW) has been widely used. However, DTW is not learnable and suffers from a trade-off between robustness against time distortion and discriminative power. In this paper, we propose a neural network model for task-adaptive time warping. Specifically, we use the attention model, called the bipartite attention model, to develop an explicit time warping mechanism with greater distortion invariance. Unlike other learnable models using DTW for warping, our model predicts all local correspondences between two time series and is trained based on metric learning, which enables it to learn the optimal data-dependent warping for the target task. We also propose to induce pre-training of our model by DTW to improve the discriminative power. Extensive experiments demonstrate the superior effectiveness of our model over DTW and its state-of-the-art performance in online signature verification.