Policy-Induced Self-Supervision Improves Representation Finetuning in Visual RL.

Sébastien M. R. Arnold Fei Sha

Published in: CoRR (2023)

Keyphrases

optimal policy
reinforcement learning
low level
neural network
visual features
visual information
action selection
state space
learning algorithm
high level
reward function
visual patterns
visual representation
markov decision process
visual properties