Incorporating Pre-training Data Matters in Unsupervised Domain Adaptation
This work addresses domain adaptation for machine learning practitioners by highlighting a previously overlooked factor, offering a novel perspective that is incremental but impactful for improving UDA methods.
The paper tackles the problem of unsupervised domain adaptation (UDA) by investigating the impact of pre-training data, revealing that target error stems from degenerative pre-trained knowledge and theoretical error bounds, and proposes TriDA, a framework that incorporates pre-training data to maintain knowledge and improve bounds, achieving state-of-the-art performance across multiple benchmarks.
In deep learning, initializing models with pre-trained weights has become the de facto practice for various downstream tasks. Many unsupervised domain adaptation (UDA) methods typically adopt a backbone pre-trained on ImageNet, and focus on reducing the source-target domain discrepancy. However, the impact of pre-training on adaptation received little attention. In this study, we delve into UDA from the novel perspective of pre-training. We first demonstrate the impact of pre-training by analyzing the dynamic distribution discrepancies between pre-training data domain and the source/ target domain during adaptation. Then, we reveal that the target error also stems from the pre-training in the following two factors: 1) empirically, target error arises from the gradually degenerative pre-trained knowledge during adaptation; 2) theoretically, the error bound depends on difference between the gradient of loss function, \ie, on the target domain and pre-training data domain. To address these two issues, we redefine UDA as a three-domain problem, \ie, source domain, target domain, and pre-training data domain; then we propose a novel framework, named TriDA. We maintain the pre-trained knowledge and improve the error bound by incorporating pre-training data into adaptation for both vanilla UDA and source-free UDA scenarios. For efficiency, we introduce a selection strategy for pre-training data, and offer a solution with synthesized images when pre-training data is unavailable during adaptation. Notably, TriDA is effective even with a small amount of pre-training or synthesized images, and seamlessly complements the two scenario UDA methods, demonstrating state-of-the-art performance across multiple benchmarks. We hope our work provides new insights for better understanding and application of domain adaptation.