EXPLAINING TIME SERIES VIA CONTRASTIVE AND LOCALLY SPARSE PERTURBATIONS
Document Type
Conference Proceeding
Publication Date
1-1-2024
Abstract
Explaining multivariate time series is a compound challenge, as it requires identifying important locations in the time series and matching complex temporal patterns. Although previous saliency-based methods addressed the challenges, their perturbation may not alleviate the distribution shift issue, which is inevitable especially in heterogeneous samples. We present ContraLSP, a locally sparse model that introduces counterfactual samples to build uninformative perturbations but keeps distribution using contrastive learning. Furthermore, we incorporate sample-specific sparse gates to generate more binary-skewed and smooth masks, which easily integrate temporal trends and select the salient features parsimoniously. Empirical studies on both synthetic and real-world datasets show that ContraLSP outperforms state-of-the-art models, demonstrating a substantial improvement in explanation quality for time series data. The source code is available at https://github.com/zichuan-liu/ContraLSP.
Identifier
85197706104 (Scopus)
Publication Title
12th International Conference on Learning Representations, ICLR 2024
Fund Ref
GABA International
Recommended Citation
Liu, Zichuan; Zhang, Yingying; Wang, Tianchun; Wang, Zefan; Luo, Dongsheng; Du, Mengnan; Wu, Min; Wang, Yi; Chen, Chunlin; Fan, Lunting; and Wen, Qingsong, "EXPLAINING TIME SERIES VIA CONTRASTIVE AND LOCALLY SPARSE PERTURBATIONS" (2024). Faculty Publications. 961.
https://digitalcommons.njit.edu/fac_pubs/961