ReLATE: Resilient Learner Selection for Multivariate Time-Series Classification Against Adversarial Attacks

2025-03-10

AI Summary

Addressing the dual challenges of high computational cost in model selection and weak adversarial robustness in multivariate time series classification, this paper proposes a similarity-driven elastic learner selection framework. Our method introduces a feature-embedding-based dynamic similarity metric across datasets, integrated with multi-model performance caching and an oracle-approximating decision strategy, enabling zero-shot reuse of adversarial-robust models without retraining. The core contribution lies in shifting model selection from per-dataset training-and-evaluation to cross-dataset transfer of robust models, thereby significantly improving both efficiency and robustness. Extensive experiments demonstrate an average 81.2% reduction in computational overhead, while maintaining classification accuracy within ±4.2% of the oracle-optimal solution under diverse adversarial attacks.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Minimizing computational overhead in time-series classification, particularly in deep learning models, presents a significant challenge. This challenge is further compounded by adversarial attacks, emphasizing the need for resilient methods that ensure robust performance and efficient model selection. We introduce ReLATE, a framework that identifies robust learners based on dataset similarity, reduces computational overhead, and enhances resilience. ReLATE maintains multiple deep learning models in well-known adversarial attack scenarios, capturing model performance. ReLATE identifies the most analogous dataset to a given target using a similarity metric, then applies the optimal model from the most similar dataset. ReLATE reduces computational overhead by an average of 81.2%, enhancing adversarial resilience and streamlining robust model selection, all without sacrificing performance, within 4.2% of Oracle.

Problem

• Minimizes computational overhead in time-series classification.
• Enhances resilience against adversarial attacks in deep learning.
• Streamlines robust model selection using dataset similarity metrics.

Innovation

• ReLATE framework enhances adversarial resilience
• Uses dataset similarity for optimal model selection
• Reduces computational overhead by 81.2%