Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Reachability-Aware Diffusion Policy Optimization
Hikmet Simsir, Kutay Demiray, Ozgur S. Oguz
Diffusion policies provide expressive action distributions for continuous-control reinforcement learning. However, safety-aware online diffusion policy optimization remains underex…
cs.LG2026
Lagrangian Perturbation Diffusion Steering: Latent Reinforcement Learning for Generative Policies
Hikmet Simsir, Ozgur S. Oguz
Behavior cloning with high-capacity generative policies achieves strong imitation performance, but is often limited by demonstration coverage and distribution shift. Direct reinfor…