Diffusion policies can be inserted into maximum-entropy RL by minimizing an upper bound on reverse KL, yielding DiffPPO, DiffSAC, and DiffWPO.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Diffusion-Augmented Markov Decision Processes for Maximum Entropy Reinforcement Learning
Diffusion policies can be inserted into maximum-entropy RL by minimizing an upper bound on reverse KL, yielding DiffPPO, DiffSAC, and DiffWPO.