Generalized Universal Domain Adaptation with Generative Flow Networks

TL;DR

This paper introduces GUDA, a new unsupervised domain adaptation problem, and proposes GFlowDA and GUAN methods leveraging generative flow networks for improved target label prediction, including unknown categories.

Contribution

It unifies different domain adaptation variants into a comprehensive framework and develops novel algorithms using generative flow networks for better exploration and adaptation.

Findings

GFlowDA outperforms existing methods on benchmark datasets.

The proposed framework effectively identifies unknown categories.

Theoretical analysis confirms the importance of exploration in GUDA.

Abstract

We introduce a new problem in unsupervised domain adaptation, termed as Generalized Universal Domain Adaptation (GUDA), which aims to achieve precise prediction of all target labels including unknown categories. GUDA bridges the gap between label distribution shift-based and label space mismatch-based variants, essentially categorizing them as a unified problem, guiding to a comprehensive framework for thoroughly solving all the variants. The key challenge of GUDA is developing and identifying novel target categories while estimating the target label distribution. To address this problem, we take advantage of the powerful exploration capability of generative flow networks and propose an active domain adaptation algorithm named GFlowDA, which selects diverse samples with probabilities proportional to a reward function. To enhance the exploration capability and effectively perceive the target label distribution, we tailor the states and rewards, and introduce an efficient solution for parent exploration and state transition. We also propose a training paradigm for GUDA called Generalized Universal Adversarial Network (GUAN), which involves collaborative optimization between GUAN and GFlowNet. Theoretical analysis highlights the importance of exploration, and extensive experiments on benchmark datasets demonstrate the superiority of GFlowDA.