Zero-shot domain adaptation based on dual-level mix and contrast

TL;DR

This paper introduces a novel zero-shot domain adaptation method that leverages dual-level mixup, domain adversarial learning, and contrastive learning to learn domain-invariant features with minimal task bias, improving performance on benchmarks.

Contribution

It proposes a new ZSDA approach combining dual-level mixup, extended domain adversarial learning, and contrastive learning to reduce task bias and enhance domain invariance.

Findings

Achieves improved performance on multiple benchmarks.

Effectively reduces task bias in domain-invariant feature learning.

Demonstrates the effectiveness of dual-level mixup and contrastive learning in ZSDA.

Abstract

Zero-shot domain adaptation (ZSDA) is a domain adaptation problem in the situation that labeled samples for a target task (task of interest) are only available from the source domain at training time, but for a task different from the task of interest (irrelevant task), labeled samples are available from both source and target domains. In this situation, classical domain adaptation techniques can only learn domain-invariant features in the irrelevant task. However, due to the difference in sample distribution between the two tasks, domain-invariant features learned in the irrelevant task are biased and not necessarily domain-invariant in the task of interest. To solve this problem, this paper proposes a new ZSDA method to learn domain-invariant features with low task bias. To this end, we propose (1) data augmentation with dual-level mixups in both task and domain to fill the absence of target task-of-interest data, (2) an extension of domain adversarial learning to learn domain-invariant features with less task bias, and (3) a new dual-level contrastive learning method that enhances domain-invariance and less task biasedness of features. Experimental results show that our proposal achieves good performance on several benchmarks.