Research map / Reinforcement learning / MDPs and planning
On the Convergence Theory of Debiased Model-Agnostic Meta-Reinforcement Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use SG-MRL instead of MAML | Meta-RL · 2D-navigation, 1 adaptation step | meta-test reward: -17.767 ± 0.106 -> -16.901 ± 0.699 | Helped |
| Use SG-MRL instead of E-MAML | Meta-RL · 2D-navigation, 1 adaptation step | meta-test reward: -17.803 ± 0.115 -> -16.901 ± 0.699 | Helped |
| Use SG-MRL instead of MAML | Meta-RL · Half-Cheetah Random Direction, 1–3 steps | meta-test reward: 465.624 ± 54.07 -> 580.143 ± 38.22; 441.247 ± 58.34 -> 580.203 ± 33.63; 477.086 ± 64.71 -> 504.747 ± 45.07 | Helped |
| Use SG-MRL instead of MAML | Meta-RL · Half-Cheetah Random Velocity, 1–2 steps | meta-test reward: -92.92 ± 0.70 -> -91.73 ± 0.34; -56.71± 6.73 -> -52.64± 6.86 | Helped |
| Use SG-MRL instead of MAML | Meta-RL · Half-Cheetah Random Velocity, 3 steps | meta-test reward: -32.48± 0.50 -> -33.39± 0.67 | Hurt |
| Use SG-MRL instead of MAML | Meta-RL · Swimmer Random Velocity, 1–3 steps | meta-test reward: 104.53 ± 24.18 -> 118.77 ± 9.99; 108.47 ± 23.36 -> 134.57 ± 1.67; 90.60 ± 14.99 -> 110.91 ± 12.56 | Helped |
Closest papers by meaning
- Taming MAML: Efficient unbiased meta-reinforcement learning (ICML 2019 · similarity 0.92)
- On the Global Optimality of Model-Agnostic Meta-Learning (ICML 2020 · similarity 0.91)
- MAMBA: an Effective World Model Approach for Meta-Reinforcement Learning (ICLR 2024 · similarity 0.90)
- Meta-Reinforcement Learning with Universal Policy Adaptation: Provable Near-Optimality under All-task Optimum Comparator (NeurIPS 2024 · similarity 0.90)
- Convergence Analysis of Policy Gradient Methods with Dynamic Stochasticity (ICML 2025 · similarity 0.88)
- Learning Optimal Deterministic Policies with Stochastic Policy Gradients (ICML 2024 · similarity 0.88)
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.