Disentangling Controllable Object through Video Prediction Improves Visual Reinforcement Learning
Year: 2020
Type: preprint
Abstract: In many vision-based reinforcement learning (RL) problems, the agent controls a movable object in its visual field, e.g., the player's avatar in video games and the robotic arm in visual grasping and manipulation. Leveraging action-conditioned video prediction, we propose an end-to-end learning framework to disentangle the controllable object from the observation signal. The disentangled representation is shown to be useful for RL as additional observation channels to the agent. Experiments on a set of Atari games with the popular Double DQN algorithm demonstrate improved sample efficiency and game performance (from 222.8% to 261.4% measured in normalized game scores, with prediction bonus reward).
Source: arXiv (Cornell University)
Authors: Zhong, Yuanyi, Schwing, Alexander, Peng, Jian
Institution: University of Illinois Urbana-Champaign
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W3007998409