Prototypical Partial Optimal Transport for Universal Domain Adaptation

TL;DR

This paper introduces m-PPOT, a novel partial optimal transport method for universal domain adaptation that effectively aligns distributions and distinguishes known from unknown samples, outperforming previous methods.

Contribution

The paper proposes a new partial distribution alignment approach called m-PPOT for UniDA, incorporating reweighted losses to improve sample classification.

Findings

Outperforms previous state-of-the-art UniDA methods on four benchmarks.

Effectively distinguishes known and unknown samples during domain adaptation.

Provides a novel partial optimal transport framework for distribution matching.

Abstract

Universal domain adaptation (UniDA) aims to transfer knowledge from a labeled source domain to an unlabeled target domain without requiring the same label sets of both domains. The existence of domain and category shift makes the task challenging and requires us to distinguish "known" samples (i.e., samples whose labels exist in both domains) and "unknown" samples (i.e., samples whose labels exist in only one domain) in both domains before reducing the domain gap. In this paper, we consider the problem from the point of view of distribution matching which we only need to align two distributions partially. A novel approach, dubbed mini-batch Prototypical Partial Optimal Transport (m-PPOT), is proposed to conduct partial distribution alignment for UniDA. In training phase, besides minimizing m-PPOT, we also leverage the transport plan of m-PPOT to reweight source prototypes and target samples, and design reweighted entropy loss and reweighted cross-entropy loss to distinguish "known" and "unknown" samples. Experiments on four benchmarks show that our method outperforms the previous state-of-the-art UniDA methods.