Active Test-Time Adaptation: Theoretical Analyses and An Algorithm

TL;DR

This paper introduces active test-time adaptation (ATTA), integrating active learning into test-time adaptation to improve performance under domain shifts, supported by theoretical analysis and a practical algorithm.

Contribution

It proposes the novel ATTA problem setting, provides theoretical guarantees, and develops the SimATTA algorithm for effective real-time sample selection.

Findings

ATTA improves test performance over traditional TTA methods.

SimATTA achieves substantial performance gains with efficiency.

Theoretical analysis confirms the benefits of limited labeled test instances.

Abstract

Test-time adaptation (TTA) addresses distribution shifts for streaming test data in unsupervised settings. Currently, most TTA methods can only deal with minor shifts and rely heavily on heuristic and empirical studies. To advance TTA under domain shifts, we propose the novel problem setting of active test-time adaptation (ATTA) that integrates active learning within the fully TTA setting. We provide a learning theory analysis, demonstrating that incorporating limited labeled test instances enhances overall performances across test domains with a theoretical guarantee. We also present a sample entropy balancing for implementing ATTA while avoiding catastrophic forgetting (CF). We introduce a simple yet effective ATTA algorithm, known as SimATTA, using real-time sample selection techniques. Extensive experimental results confirm consistency with our theoretical analyses and show that the proposed ATTA method yields substantial performance improvements over TTA methods while maintaining efficiency and shares similar effectiveness to the more demanding active domain adaptation (ADA) methods. Our code is available at https://github.com/divelab/ATTA