Progressive Graph Learning for Open-Set Domain Adaptation

TL;DR

This paper introduces a novel Progressive Graph Learning framework for open-set domain adaptation, effectively handling additional classes in target data and outperforming existing methods through an end-to-end graph neural network approach.

Contribution

It proposes an end-to-end PGL framework combining graph neural networks and adversarial learning for open-set domain adaptation, improving target error bounds.

Findings

Outperforms state-of-the-art methods on three open-set benchmarks.

Achieves a tighter upper bound of target error.

Effectively suppresses conditional shift in open-set scenarios.

Abstract

Domain shift is a fundamental problem in visual recognition which typically arises when the source and target data follow different distributions. The existing domain adaptation approaches which tackle this problem work in the closed-set setting with the assumption that the source and the target data share exactly the same classes of objects. In this paper, we tackle a more realistic problem of open-set domain shift where the target data contains additional classes that are not present in the source data. More specifically, we introduce an end-to-end Progressive Graph Learning (PGL) framework where a graph neural network with episodic training is integrated to suppress underlying conditional shift and adversarial learning is adopted to close the gap between the source and target distributions. Compared to the existing open-set adaptation approaches, our approach guarantees to achieve a tighter upper bound of the target error. Extensive experiments on three standard open-set benchmarks evidence that our approach significantly outperforms the state-of-the-arts in open-set domain adaptation.