Songyuan Zhang
@songyuan-zhang · 1 works
Researcher working on safe reinforcement learning and safety filters for AI systems.
Recent posts by Songyuan Zhang
- FAITH: Feasibility-Aware Safety-Filtered RL for High-Dimensional Systems - Safe reinforcement learning commonly places safety and task performance in the same policy objective, where they can introduce competing updates. Safety filters separate them at action execution,… · Paper · arxiv.org