Research map / Reinforcement learning / Offline RL
Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamics
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 |
|---|---|---|---|
| Add Koopman-derived symmetry shifts to CQL training data | CQL · D4RL tasks overall | normalized return | Mixed |
| Add Koopman-derived symmetry shifts to CQL training data | CQL · antmaze-medium-diverse | normalized return: 82.3 -> 90.5 | Helped |
| Add Koopman-derived symmetry shifts to CQL training data | CQL · hopper-medium | normalized return: 81.3 -> 94.2 | Helped |
| Add Koopman-derived symmetry shifts to CQL training data | CQL · kitchen-partial | normalized return: 83.6 -> 95.9 | Helped |
| Add Koopman-derived symmetry shifts to CQL training data | CQL · cheetah-random | normalized return: 53.9 -> 49.2 | Hurt |
| Add Koopman-derived symmetry shifts to CQL training data | CQL · walker-random | normalized return: 25.1 -> 17.6 | Hurt |
| Use random-action offline data rather than more structured behavior data | KFC · D4RL -random splits | normalized return | Hurt |
Closest papers by meaning
- Efficient Dynamics Modeling in Interactive Environments with Koopman Theory (ICLR 2024 · similarity 0.90)
- Inverting the Bellman Equation: From $Q$-Values to World Models (NeurIPS 2026 · similarity 0.88)
- Learning on One Mode: Addressing Multi-modality in Offline Reinforcement Learning (ICLR 2025 · similarity 0.88)
- Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models (NeurIPS 2025 · similarity 0.88)
- Latent Mixture of Symmetries for Sample-Efficient Dynamic Learning (NeurIPS 2025 · similarity 0.87)
- Reasoning with Latent Diffusion in Offline Reinforcement Learning (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.