Augmented Contrastive Clustering with Uncertainty-Aware Prototyping for Time Series Test Time Adaptation

TL;DR

This paper introduces ACCUP, a novel test-time adaptation method for time series data that uses augmentation, uncertainty-aware prototypes, and contrastive clustering to improve model performance without labels.

Contribution

The paper presents a new TTA approach specifically designed for time series, incorporating augmentation ensemble, uncertainty-aware prototypes, and contrastive clustering, addressing limitations of visual-focused methods.

Findings

Effective on three real-world time series datasets

Improves feature discriminability and clustering quality

Demonstrates generalization to visual data

Abstract

Test-time adaptation aims to adapt pre-trained deep neural networks using solely online unlabelled test data during inference. Although TTA has shown promise in visual applications, its potential in time series contexts remains largely unexplored. Existing TTA methods, originally designed for visual tasks, may not effectively handle the complex temporal dynamics of real-world time series data, resulting in suboptimal adaptation performance. To address this gap, we propose Augmented Contrastive Clustering with Uncertainty-aware Prototyping (ACCUP), a straightforward yet effective TTA method for time series data. Initially, our approach employs augmentation ensemble on the time series data to capture diverse temporal information and variations, incorporating uncertainty-aware prototypes to distill essential characteristics. Additionally, we introduce an entropy comparison scheme to selectively acquire more confident predictions, enhancing the reliability of pseudo labels. Furthermore, we utilize augmented contrastive clustering to enhance feature discriminability and mitigate error accumulation from noisy pseudo labels, promoting cohesive clustering within the same class while facilitating clear separation between different classes. Extensive experiments conducted on three real-world time series datasets and an additional visual dataset demonstrate the effectiveness and generalization potential of the proposed method, advancing the underexplored realm of TTA for time series data.