Research map / Reinforcement learning / Visual and representation RL
Reinforcement Learning with Fast and Forgetful Memory
Steven Morad, Ryan Kortvelesy, Stephan Liwicki, Amanda Prorok
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 recurrent memory with FFM | recurrent RL models · POPGym | MMER: 0.349 -> 0.400 | Improved |
| Add input gating (ablation) | FFM · POPGym tasks | mean reward (μ): 0.392 -> 0.399 | Improved |
| Add output gating (ablation) | FFM · POPGym tasks | mean reward (μ): 0.395 -> 0.399 | Improved |
| Add temporal context (ablation) | FFM · POPGym tasks | mean reward (μ): 0.382 -> 0.399 | Improved |
| Learn context rather than fixing it (ablation) | FFM · POPGym tasks | mean reward (μ): 0.383 -> 0.399 | Improved |
| Add decay (forgetting) (ablation) | FFM · POPGym tasks | mean reward (μ): 0.333 -> 0.399 | Improved |
| Learn decay rather than fixing it (ablation) | FFM · POPGym tasks | mean reward (μ): 0.390 -> 0.399 | Improved |
| Use the outer product rather than the Hadamard product for γ (ablation) | FFM · POPGym tasks | mean reward (μ): 0.395 -> 0.399 | Improved |
Closest papers by meaning
- Recurrent Reinforcement Learning with Memoroids (NeurIPS 2024 · similarity 0.90)
- Stable Hadamard Memory: Revitalizing Memory-Augmented Agents for Reinforcement Learning (ICLR 2025 · similarity 0.88)
- A Forget-and-Grow Strategy for Deep Reinforcement Learning Scaling in Continuous Control (ICML 2025 · similarity 0.88)
- Efficient Recurrent Off-Policy RL Requires a Context-Encoder-Specific Learning Rate (NeurIPS 2024 · similarity 0.88)
- Mastering Memory Tasks with World Models (ICLR 2024 · similarity 0.87)
- Reasoning Under 1 Billion: Memory-Augmented Reinforcement Learning for Large Language Models (NeurIPS 2026 · 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.
