Research map / Reinforcement learning / Policy gradient and actor-critic
OFFER: Off-Environment 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 |
|---|---|---|---|
| learn a proposal distribution over environment variables (OFFER) vs primary optimisation only | policy gradient · Unaliased Mountain Car | average per-episode penalty (lower better) | Helped |
| learn a proposal distribution over environment variables (OFFER) | policy gradient · Aliased Mountain Car | average per-episode penalty (lower better) | Helped |
| learn 4-feature proposal distribution (OFFER) | policy gradient · 4-Featured Mountain Car | average per-episode penalty (lower better) | Helped |
| learn proposal distribution (OFFER) | policy gradient · Robotic Arm | task penalty (lower better) | Helped |
| learn proposal distribution (OFFER), on normal events only | policy gradient · Unaliased Mountain Car, normal events | penalty on normal events | Hurt |
| learn proposal distribution (OFFER), on normal events with shared features | policy gradient · Aliased Mountain Car, normal events | penalty on normal events | No clear effect |
Closest papers by meaning
- Policy-conditioned Environment Models are More Generalizable (ICML 2024 · similarity 0.88)
- Model Selection for Off-policy Evaluation: New Algorithms and Experimental Protocol (NeurIPS 2025 · similarity 0.87)
- Environment Design for Inverse Reinforcement Learning (ICML 2024 · similarity 0.87)
- OCEAN-MBRL: Offline Conservative Exploration for Model-Based Offline Reinforcement Learning (AAAI 2024 · similarity 0.87)
- Preference Elicitation for Offline Reinforcement Learning (ICLR 2025 · similarity 0.87)
- Skill or Luck? Return Decomposition via Advantage Functions (ICLR 2024 · similarity 0.87)
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.