构建统一基准框架，评估强化学习在对抗攻击下的鲁棒性
RoAd-RL: A Unified Library and Benchmark for Robust Adversarial Reinforcement Learning
- 提供统一接口支持策略、攻击、防御与度量的标准化实现
- 在192种配置下测试发现部分防御反而削弱性能
- 适合研究对抗性强化学习的开发者和安全评估人员
深度强化学习在机器人和自主系统中取得显著进展，但仍易受对抗扰动影响，导致性能严重下降。现有研究受限于实现碎片化、评估协议不一致和可复现性差。为此，我们提出开源基准框架 RoAd-RL，提供政策、攻击、防御与鲁棒性度量的统一抽象，并支持与 Stable-Baselines3 和 Gymnasium 的无缝集成。我们在 LunarLander 与 Highway-v0 环境中对 DQN、PPO、SAC 三种智能体在 192 种攻击-防御组合下进行评估。结果表明不同环境间鲁棒性差异显著，部分常用防御甚至比攻击本身更具破坏性，而时间平滑方法始终表现优异。RoAd-RL 为对抗强化学习研究建立了标准化基准，项目已公开发布于 https://pypi.org/project/road-rl。
原文摘要 · Abstract (English)
Deep Reinforcement Learning (DRL) has achieved significant success in robotics and autonomous systems, yet remains vulnerable to adversarial perturbations that can severely degrade performance. Research in adversarial reinforcement learning is often limited by fragmented implementations, inconsistent evaluation protocols, and poor reproducibility. To address these challenges, we present \textbf{RoAd-RL}, an open-source benchmarking framework that provides unified abstractions for policies, attacks, defenses, and robustness metrics, together with reproducible evaluation pipelines and seamless integration with Stable-Baselines3 and Gymnasium. We evaluate DQN, PPO, and SAC agents in LunarLander and Highway-v0 under 192 attack-defense configurations. Results reveal substantial variations in robustness across environments and show that some commonly used defenses can be more detrimental than the attacks they aim to mitigate, while temporal smoothing consistently achieves strong performance. RoAd-RL establishes a standardized benchmark for adversarial reinforcement learning research and is publicly available at https://pypi.org/project/road-rl.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。