DyCVAE: Learning Dynamic Causal Factors for Non-Stationary Series Domain Generalization (Student Abstract)

Weifeng Zhang, Zhiyuan Wang, Kunpeng Zhang, Ting Zhong, Fan Zhou

AAAI 2023 pp. 16382-16383

doi:10.1609/AAAI.V37I13.27051 /aaai/2023/zhang2023aaai-dycvae/

Abstract

Learning domain-invariant representations is a major task of out-of-distribution generalization. To address this issue, recent efforts have taken into accounting causality, aiming at learning the causal factors with regard to tasks. However, extending existing generalization methods for adapting non-stationary time series may be ineffective, because they fail to model the underlying causal factors due to temporal-domain shifts except for source-domain shifts, as pointed out by recent studies. To this end, we propose a novel model DyCVAE to learn dynamic causal factors. The results on synthetic and real datasets demonstrate the effectiveness of our proposed model for the task of generalization in time series domain.

PDF AAAI Semantic Scholar

Cite

Text

Zhang et al. "DyCVAE: Learning Dynamic Causal Factors for Non-Stationary Series Domain Generalization (Student Abstract)." AAAI Conference on Artificial Intelligence, 2023. doi:10.1609/AAAI.V37I13.27051

Markdown

[Zhang et al. "DyCVAE: Learning Dynamic Causal Factors for Non-Stationary Series Domain Generalization (Student Abstract)." AAAI Conference on Artificial Intelligence, 2023.](https://mlanthology.org/aaai/2023/zhang2023aaai-dycvae/) doi:10.1609/AAAI.V37I13.27051

BibTeX

@inproceedings{zhang2023aaai-dycvae,
  title     = {{DyCVAE: Learning Dynamic Causal Factors for Non-Stationary Series Domain Generalization (Student Abstract)}},
  author    = {Zhang, Weifeng and Wang, Zhiyuan and Zhang, Kunpeng and Zhong, Ting and Zhou, Fan},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2023},
  pages     = {16382-16383},
  doi       = {10.1609/AAAI.V37I13.27051},
  url       = {https://mlanthology.org/aaai/2023/zhang2023aaai-dycvae/}
}