AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Leveraging Explanation to Improve Generalization of Meta Reinforcement Learning

Shicheng Liu, Minghui Zhu

ICLR 2026 · Reinforcement learning › Visual and representation RL · approach: Task-inference meta-RL

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
add XMRL critical-task augmentationMAML · Drone navigationsuccess rate: 0.87 -> 0.97Improved
add XMRL critical-task augmentationMAML · Stock marketcumulative reward: 359.13 -> 421.13Improved
add XMRL critical-task augmentationMAML · HalfCheetahcumulative reward: -68.89 -> -44.67Improved
add XMRL critical-task augmentationMAML · Antcumulative reward: 100.64 -> 119.15Improved
add task weighting instead of augmentationMAML · Drone navigationsuccess rate: 0.87 -> 0.87No clear effect
add meta augmentation with predefined mixing distribution instead of learned oneMAML · Stock marketcumulative reward: 359.13 -> 389.17Improved
add meta regularization instead of XMRLMAML · Antcumulative reward: 100.64 -> 104.15Improved
add XMRL, performance on poorly adapted tasks (ablation)MAML · Drone navigation, poorly adapted taskssuccess rate: 0.55 -> 0.93Improved
add XMRL, performance on poorly adapted tasks (ablation)MAML · Stock market, poorly adapted taskscumulative reward: 71.05 -> 381.33Improved
add XMRL, performance on poorly adapted tasks (ablation)MAML · HalfCheetah, poorly adapted taskscumulative reward: -162.09 -> -55.00Improved
add XMRL, performance on non-poorly adapted tasks (ablation)MAML · Stock market, non-poorly adapted taskscumulative reward: 431.15 -> 431.08No clear effect
add XMRL, performance on non-poorly adapted tasks (ablation)MAML · Ant, non-poorly adapted taskscumulative reward: 115.88 -> 124.02Improved

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