Self-Supervised Noisy Label Learning for Source-Free Unsupervised Domain Adaptation

TL;DR

This paper introduces a novel self-supervised noisy label learning approach for source-free unsupervised domain adaptation, enabling effective model fine-tuning using pre-generated and self-generated labels without access to source data.

Contribution

It proposes a new method combining self-supervised learning with noisy label handling for source-free domain adaptation, achieving state-of-the-art results.

Findings

Achieves superior performance over existing methods.

Effectively utilizes pre-trained models without source data.

Surpasses previous state-of-the-art by a large margin.

Abstract

It is a strong prerequisite to access source data freely in many existing unsupervised domain adaptation approaches. However, source data is agnostic in many practical scenarios due to the constraints of expensive data transmission and data privacy protection. Usually, the given source domain pre-trained model is expected to optimize with only unlabeled target data, which is termed as source-free unsupervised domain adaptation. In this paper, we solve this problem from the perspective of noisy label learning, since the given pre-trained model can pre-generate noisy label for unlabeled target data via directly network inference. Under this problem modeling, incorporating self-supervised learning, we propose a novel Self-Supervised Noisy Label Learning method, which can effectively fine-tune the pre-trained model with pre-generated label as well as selfgenerated label on the fly. Extensive experiments had been conducted to validate its effectiveness. Our method can easily achieve state-of-the-art results and surpass other methods by a very large margin. Code will be released.