On the Power of Pre-training for Generalization in RL: Provable Benefits and Hardness.

Haotian Ye Xiaoyu Chen Liwei Wang Simon Shaolei Du

Published in: ICML (2023)

Keyphrases

reinforcement learning
test set
power consumption
learning machines
training set
state space
multi agent
np complete
training examples
function approximation
training process
computational complexity
phase transition
training error
partially observable domains