Adversarial Reinforcement Learning for Unsupervised Domain Adaptation

Transferring knowledge from an existing labeled domain to a new domain often suffers from domain shift in which performance degrades because of differences between the domains. Domain adaptation has been a prominent method to mitigate such a problem. There have been many pretrained neural networks for feature extraction. However, little work discusses how to select the best feature instances across different pre-trained models for both the source and target domain. We propose a novel approach to select features by employing reinforcement learning, which learns to select the most relevant features across two domains. Specifically, in this framework, we employ Q-learning to learn policies for an agent to make feature selection decisions by approximating the action-value function. After selecting the best features, we propose an adversarial distribution alignment learning to improve the prediction results. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art methods.

Automated summary

The article mainly discusses how to solve the problem of domain shift by selecting the most relevant features through reinforcement learning, and proposes adversarial distribution alignment learning to improve prediction results. Extensive experiments show that the proposed method outperforms existing state-of-the-art methods.