DMBP: Diffusion model-based predictor for robust offline reinforcement learning against state observation perturbations.

Zhihe Yang Yunjian Xu

Published in: ICLR (2024)

Keyphrases

reinforcement learning
state space
model free
real time
optimal policy
multi agent systems
markov decision processes
function approximation
reinforcement learning algorithms
transition model
optimal control
anisotropic diffusion
robust estimation