Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Neural Episodic Control
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 |
|---|---|---|---|
| Use NEC instead of Prioritised Replay | Atari agents · Atari 57 games, 10M frames (median across games) | human-normalised score: 22.4% -> 54.6% | Helped |
| Use NEC instead of Prioritised Replay | Atari agents · Atari 57 games, 40M frames (median across games) | human-normalised score: 89.0% -> 83.3% | Hurt |
| Use NEC instead of Prioritised Replay | Atari agents · Atari 57 games, 10M frames (mean across games) | human-normalised score: 71.5% -> 99.8% | Helped |
| Use NEC instead of Q*(λ) | Atari agents · Atari 57 games, 40M frames (mean across games) | human-normalised score: 440.9% -> 144.8% | Hurt |
| Use a reward-trained embedding for value interpolation instead of MFEC's reward-independent embedding | NEC · Atari 57 games, 10M frames (median across games) | human-normalised score: 45.4% -> 54.6% | Helped |
| Avoid reward clipping | NEC compared with clipped-reward parametric methods · Atari games with rewards outside [-1, 1], including Bowling, Frostbite, H.E.R.O., Ms. Pac-Man, and Alien | score | Helped |
| Use NEC instead of other algorithms | Atari agents · Pong and Boxing | learning speed | Helped |
Closest papers by meaning
- Model-Based Episodic Memory Induces Dynamic Hybrid Controls (NeurIPS 2021 · similarity 0.90)
- Fast deep reinforcement learning using online adjustments from the past (NeurIPS 2018 · similarity 0.89)
- A Forget-and-Grow Strategy for Deep Reinforcement Learning Scaling in Continuous Control (ICML 2025 · similarity 0.86)
- Overcoming Slow Decision Frequencies in Continuous Control: Model-Based Sequence Reinforcement Learning for Model-Free Control (ICLR 2025 · similarity 0.86)
- Relating Hopfield Networks to Episodic Control (NeurIPS 2024 · similarity 0.86)
- Efficient Episodic Memory Utilization of Cooperative Multi-Agent Reinforcement Learning (ICLR 2024 · 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.