Research map / Reinforcement learning / Visual and representation RL
Dynamic allocation of limited memory resources in reinforcement learning
Nisheet Patel, Luigi Acerbi, Alexandre Pouget
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 |
|---|---|---|---|
| allocate memory precision flexibly (DRA) instead of equal precision for all memories (ablation) | DRA · 2D grid-world | objective: 2× improvement | Improved |
| set lambda to 0 so DRA reduces to SARSA (ablation) | DRA · 2D grid-world | episodes to converge: 1.5× | Mixed |
| allocate memory precision flexibly instead of equal precision (ablation) | DRA · Mountain car | objective: 1.3× improvement | Improved |
| set lambda to 0 baseline (ablation) | DRA · Mountain car | episodes to converge: 1.3× | Mixed |
| start with more memory resources (lower initial sigma0) | DRA · 2D grid-world | learning speed | Improved |
| use advantage gradient instead of alternative gradients (Eq. 4) (ablation) | DRA · Mountain car | objective | Improved |
| increase search budget toward 100% | DRA vs equal-precision · Planning task | performance advantage of flexible allocation | Hurt |
Closest papers by meaning
- Efficient Allocation of Working Memory Resource for Utility Maximization in Humans and Recurrent Neural Networks (NeurIPS 2025 · similarity 0.89)
- Mem-T: Densifying Rewards for Long-Horizon Memory Agents (ICML 2026 · similarity 0.85)
- Deep Reinforcement Learning with Time-Scale Invariant Memory (AAAI 2025 · similarity 0.85)
- Mitigating Partial Observability in Sequential Decision Processes via the Lambda Discrepancy (NeurIPS 2024 · similarity 0.85)
- Meta-Cognitive Memory Policy Optimization for Long-Horizon LLM Agents (NeurIPS 2026 · similarity 0.85)
- Stable Hadamard Memory: Revitalizing Memory-Augmented Agents for Reinforcement Learning (ICLR 2025 · similarity 0.85)
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.
