Memory-Assisted Sub-Prototype Mining for Universal Domain Adaptation

TL;DR

This paper introduces MemSPM, a memory-assisted method for universal domain adaptation that learns sub-classes within categories to handle concept shifts, improving transferability and achieving state-of-the-art results.

Contribution

The paper proposes a novel Memory-Assisted Sub-Prototype Mining approach to learn intra-class differences and sub-classes, addressing semantic ambiguity and improving adaptation performance.

Findings

Achieves state-of-the-art performance on multiple benchmarks.

Effectively models intra-class differences in domain adaptation.

Enhances transferability by learning sub-classes within categories.

Abstract

Universal domain adaptation aims to align the classes and reduce the feature gap between the same category of the source and target domains. The target private category is set as the unknown class during the adaptation process, as it is not included in the source domain. However, most existing methods overlook the intra-class structure within a category, especially in cases where there exists significant concept shift between the samples belonging to the same category. When samples with large concept shift are forced to be pushed together, it may negatively affect the adaptation performance. Moreover, from the interpretability aspect, it is unreasonable to align visual features with significant differences, such as fighter jets and civil aircraft, into the same category. Unfortunately, due to such semantic ambiguity and annotation cost, categories are not always classified in detail, making it difficult for the model to perform precise adaptation. To address these issues, we propose a novel Memory-Assisted Sub-Prototype Mining (MemSPM) method that can learn the differences between samples belonging to the same category and mine sub-classes when there exists significant concept shift between them. By doing so, our model learns a more reasonable feature space that enhances the transferability and reflects the inherent differences among samples annotated as the same category. We evaluate the effectiveness of our MemSPM method over multiple scenarios, including UniDA, OSDA, and PDA. Our method achieves state-of-the-art performance on four benchmarks in most cases.