We Need a More Robust Classifier: Dual Causal Learning Empowers Domain-Incremental Time Series Classification

2026-01-15

AI Summary

This work addresses the vulnerability of existing time series classification methods in domain-incremental learning, where models are often misled by confounding intra-class and inter-class features, resulting in poor robustness. To mitigate this, we propose DualCD, a lightweight dual causal disentanglement framework that separates causal from spurious features through temporal feature disentanglement. DualCD introduces a novel dual causal intervention mechanism that constructs counterfactual variants to compel the model to rely solely on causal features for prediction. The framework is designed to be seamlessly integrated into mainstream time series classifiers and establishes the first comprehensive benchmark for domain-incremental time series classification. Extensive experiments across multiple datasets and backbone architectures demonstrate that DualCD significantly enhances both robustness and generalization, validating its effectiveness and broad applicability.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Knowledge Representation and Reasoning: Action, Change, and Causality

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

The World Wide Web thrives on intelligent services that rely on accurate time series classification, which has recently witnessed significant progress driven by advances in deep learning. However, existing studies face challenges in domain incremental learning. In this paper, we propose a lightweight and robust dual-causal disentanglement framework (DualCD) to enhance the robustness of models under domain incremental scenarios, which can be seamlessly integrated into time series classification models. Specifically, DualCD first introduces a temporal feature disentanglement module to capture class-causal features and spurious features. The causal features can offer sufficient predictive power to support the classifier in domain incremental learning settings. To accurately capture these causal features, we further design a dual-causal intervention mechanism to eliminate the influence of both intra-class and inter-class confounding features. This mechanism constructs variant samples by combining the current class's causal features with intra-class spurious features and with causal features from other classes. The causal intervention loss encourages the model to accurately predict the labels of these variant samples based solely on the causal features. Extensive experiments on multiple datasets and models demonstrate that DualCD effectively improves performance in domain incremental scenarios. We summarize our rich experiments into a comprehensive benchmark to facilitate research in domain incremental time series classification.

Problem

• domain-incremental learning
• time series classification
• robustness
• causal features
• distribution shift

Innovation

• dual-causal disentanglement
• domain-incremental learning
• time series classification
• causal intervention
• feature disentanglement