Safety-Gymnasium: A Unified Safe Reinforcement Learning Benchmark
This benchmark addresses safety concerns in AI deployment by providing tools for the research community to develop safer reinforcement learning for real-world applications, though it is incremental as it builds on existing SafeRL work.
The paper introduces Safety-Gymnasium, a unified benchmark for safe reinforcement learning that includes safety-critical tasks for single and multi-agent scenarios with vector and vision-only inputs, along with a library of 16 state-of-the-art SafeRL algorithms to facilitate evaluation and comparison.
Artificial intelligence (AI) systems possess significant potential to drive societal progress. However, their deployment often faces obstacles due to substantial safety concerns. Safe reinforcement learning (SafeRL) emerges as a solution to optimize policies while simultaneously adhering to multiple constraints, thereby addressing the challenge of integrating reinforcement learning in safety-critical scenarios. In this paper, we present an environment suite called Safety-Gymnasium, which encompasses safety-critical tasks in both single and multi-agent scenarios, accepting vector and vision-only input. Additionally, we offer a library of algorithms named Safe Policy Optimization (SafePO), comprising 16 state-of-the-art SafeRL algorithms. This comprehensive library can serve as a validation tool for the research community. By introducing this benchmark, we aim to facilitate the evaluation and comparison of safety performance, thus fostering the development of reinforcement learning for safer, more reliable, and responsible real-world applications. The website of this project can be accessed at https://sites.google.com/view/safety-gymnasium.