AI Paper Atlas

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

NeurIPS 2023 · Reinforcement learning › Policy gradient and actor-critic · approach: Model-based RL with guarantees · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
apply RoML on top of VariBADVariBAD · HalfCheetah-VelCVaR0.05 return: -202 -> -184Improved
apply RoML on top of VariBADVariBAD · HalfCheetah-MassCVaR0.05 return: 1072 -> 1259Improved
apply RoML on top of VariBADVariBAD · HalfCheetah-BodyCVaR0.05 return: 835 -> 935Improved
apply RoML on top of VariBADVariBAD · Humanoid-MassCVaR0.05 return: 1283 -> 1378Improved
apply RoML on top of VariBADVariBAD · Ant-GoalCVaR0.05 return: -500 -> -454Improved
apply RoML on top of PEARLPEARL · HalfCheetah-VelCVaR0.05 return: -1156 -> -1089Improved
apply RoML on top of PEARLPEARL · HalfCheetah-MassCVaR0.05 return: 1115 -> 1186Improved
apply RoML on top of VariBADVariBAD · HalfCheetah-Massmean return: 1558 -> 1581No clear effect
apply RoML on top of VariBADVariBAD · Humanoid-Massmean return: 1645 -> 1580Hurt
apply RoML on top of VariBADVariBAD · HalfCheetah-Velmean return: -82 -> -95Hurt
use naive CVaR-ML instead of RoMLCVaR-ML · HalfCheetah-MassCVaR0.05 return: 1259 -> 38Hurt
replace CEM with naive adversarial sampler (ablation)RoML · MuJoCoCVaR returnHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy