Research map / Reinforcement learning / Policy gradient and actor-critic
Train Hard, Fight Easy: Robust Meta Reinforcement Learning
Ido Greenberg, Shie Mannor, Gal Chechik, Eli Meirom
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| apply RoML on top of VariBAD | VariBAD · HalfCheetah-Vel | CVaR0.05 return: -202 -> -184 | Improved |
| apply RoML on top of VariBAD | VariBAD · HalfCheetah-Mass | CVaR0.05 return: 1072 -> 1259 | Improved |
| apply RoML on top of VariBAD | VariBAD · HalfCheetah-Body | CVaR0.05 return: 835 -> 935 | Improved |
| apply RoML on top of VariBAD | VariBAD · Humanoid-Mass | CVaR0.05 return: 1283 -> 1378 | Improved |
| apply RoML on top of VariBAD | VariBAD · Ant-Goal | CVaR0.05 return: -500 -> -454 | Improved |
| apply RoML on top of PEARL | PEARL · HalfCheetah-Vel | CVaR0.05 return: -1156 -> -1089 | Improved |
| apply RoML on top of PEARL | PEARL · HalfCheetah-Mass | CVaR0.05 return: 1115 -> 1186 | Improved |
| apply RoML on top of VariBAD | VariBAD · HalfCheetah-Mass | mean return: 1558 -> 1581 | No clear effect |
| apply RoML on top of VariBAD | VariBAD · Humanoid-Mass | mean return: 1645 -> 1580 | Hurt |
| apply RoML on top of VariBAD | VariBAD · HalfCheetah-Vel | mean return: -82 -> -95 | Hurt |
| use naive CVaR-ML instead of RoML | CVaR-ML · HalfCheetah-Mass | CVaR0.05 return: 1259 -> 38 | Hurt |
| replace CEM with naive adversarial sampler (ablation) | RoML · MuJoCo | CVaR return | Hurt |
Closest papers by meaning
- Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learning (ICLR 2025 · similarity 0.91)
- Distributionally Robust Reinforcement Learning with Interactive Data Collection: Fundamental Hardness and Near-Optimal Algorithms (NeurIPS 2024 · similarity 0.91)
- Time-Constrained Robust MDPs (NeurIPS 2024 · similarity 0.89)
- Adversarial Diffusion for Robust Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
- MAMBA: an Effective World Model Approach for Meta-Reinforcement Learning (ICLR 2024 · similarity 0.89)
- Empirical Study on Robustness and Resilience in Cooperative Multi-Agent Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
