Research map / Reinforcement learning / Visual and representation RL
Structured State Space Models for In-Context Reinforcement Learning
Chris Lu, Yannick Schroecker, Albert Gu, Emilio Parisotto, Jakob Nicolaus Foerster, Satinder Singh, Feryal Behbahani
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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace GRU with S5 (reset-capable) recurrent layer | recurrent PPO agent · POPGym Repeat Previous Hard | MMER: -0.46 -> 0.91 | Improved |
| replace GRU with S5 | recurrent PPO agent · POPGym Noisy Stateless Pendulum Hard | MMER: 0.61 -> 0.55 | Hurt |
| replace GRU with S5 | recurrent PPO agent · POPGym Stateless Pendulum Hard | MMER: 0.75 -> 0.79 | Improved |
| replace GRU with S5 | recurrent PPO agent · POPGym Noisy Stateless CartPole Hard | MMER: 0.27 -> 0.28 | No clear effect |
| replace GRU with S5 | recurrent PPO agent · POPGym runtime | speedup | Improved |
| replace GRU with S5 on randomly-projected StatelessCartPole meta-RL | RL2 agent · Randomly-projected StatelessCartPole | return and runtime | Improved |
| replace LSTM with S5 | Muesli meta-RL agent · Multi-environment DMControl meta-RL (train) | return | Improved |
Closest papers by meaning
- Overcoming Slow Decision Frequencies in Continuous Control: Model-Based Sequence Reinforcement Learning for Model-Free Control (ICLR 2025 · similarity 0.88)
- XLand-100B: A Large-Scale Multi-Task Dataset for In-Context Reinforcement Learning (ICLR 2025 · similarity 0.88)
- Balancing Context Length and Mixing Times for Reinforcement Learning at Scale (NeurIPS 2024 · similarity 0.87)
- Zero-shot Model-based Reinforcement Learning using Large Language Models (ICLR 2025 · similarity 0.87)
- Drama: Mamba-Enabled Model-Based Reinforcement Learning Is Sample and Parameter Efficient (ICLR 2025 · similarity 0.86)
- Decision Mamba: A Multi-Grained State Space Model with Self-Evolution Regularization for Offline RL (NeurIPS 2024 · similarity 0.86)
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.
