Safety Gymnasium
PulseAugur coverage of Safety Gymnasium — every cluster mentioning Safety Gymnasium across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
EpiFlow advances safe offline reinforcement learning by balancing performance and safety
Researchers have developed EpiFlow, a novel approach to offline reinforcement learning that aims to balance safety and performance in critical applications. Unlike previous methods that often compromise safety for bette…
-
New research proposes CVaR to measure tail-end safety in reinforcement learning
A new research paper introduces a method to better evaluate the safety of reinforcement learning policies by examining the tail end of episodic costs, not just the average. The study proposes using Conditional Value at …
-
New algorithm synthesizes formally verified control policies for autonomous systems
Researchers have developed SMC-ES, a new algorithm that integrates Evolutionary Strategies with Statistical Model Checking to automatically synthesize control policies for autonomous systems. This method provides formal…
-
New research advances AI alignment and sample efficiency in imitation learning
Researchers have developed new methods to improve the alignment of artificial intelligence agents with human values. One approach, Feedback Manipulation Regularization (FMR), uses evaluative feedback as a corrective sig…
-
New RL methods enhance LLM training stability and efficiency · 7 sources tracked
Researchers have developed several new methods to improve the stability and efficiency of reinforcement learning (RL) in large language models (LLMs). STARE addresses policy entropy collapse by reweighting token-level a…
-
LLMs help train safer AI policies with limited violation data
Researchers have developed PROCO, a novel framework for offline safe reinforcement learning designed for scenarios with limited violation data. This model-based approach integrates natural language knowledge from large …