UBR$^2$S: Uncertainty-Based Resampling and Reweighting Strategy for Unsupervised Domain Adaptation

TL;DR

UBR$^2$S introduces an uncertainty-based resampling and reweighting strategy using Monte Carlo dropout to improve unsupervised domain adaptation, achieving state-of-the-art results across multiple datasets.

Contribution

The paper presents a novel uncertainty-driven approach for dynamic pseudo-label resampling and reweighting in UDA, enhancing adaptation performance.

Findings

Achieves state-of-the-art results on multiple UDA datasets.

Effective in both single-source and multi-source adaptation.

Applicable to various network architectures.

Abstract

Unsupervised domain adaptation (UDA) deals with the adaptation process of a model to an unlabeled target domain while annotated data is only available for a given source domain. This poses a challenging task, as the domain shift between source and target instances deteriorates a model's performance when not addressed. In this paper, we propose UBR$^2$S - the Uncertainty-Based Resampling and Reweighting Strategy - to tackle this problem. UBR$^2$S employs a Monte Carlo dropout-based uncertainty estimate to obtain per-class probability distributions, which are then used for dynamic resampling of pseudo-labels and reweighting based on their sample likelihood and the accompanying decision error. Our proposed method achieves state-of-the-art results on multiple UDA datasets with single and multi-source adaptation tasks and can be applied to any off-the-shelf network architecture. Code for our method is available at https://gitlab.com/tringwald/UBR2S.