Visual Transfer for Reinforcement Learning via Wasserstein Domain Confusion [article]

Josh Roy, George Konidaris
2020 arXiv   pre-print
We introduce Wasserstein Adversarial Proximal Policy Optimization (WAPPO), a novel algorithm for visual transfer in Reinforcement Learning that explicitly learns to align the distributions of extracted features between a source and target task. WAPPO approximates and minimizes the Wasserstein-1 distance between the distributions of features from source and target domains via a novel Wasserstein Confusion objective. WAPPO outperforms the prior state-of-the-art in visual transfer and successfully
more » ... transfers policies across Visual Cartpole and two instantiations of 16 OpenAI Procgen environments.
arXiv:2006.03465v1 fatcat:yfobpv4tdrhp5njilmgygf5qnu