Research map / Reinforcement learning / Visual and representation RL
Learning to Play No-Press Diplomacy with Best Response Policy Iteration
Thomas Anthony, Tom Eccles, Andrea Tacchetti, János Kramár, Ian Gemp, Thomas Hudson, Nicolas Porcel, Marc Lanctot, Julien Perolat, Richard Everett, Satinder P. Singh, Thore Graepel and 1 more
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 |
|---|---|---|---|
| add Sampled Best Response policy iteration (FPPI-2) on top of supervised imitation | SL (ours) · No-Press Diplomacy 1v6 vs Albert | average score: 36.3% -> 68.7% | Improved |
| use BRPI instead of A2C RL fine-tuning of imitation agent | DipNet · No-Press Diplomacy 1v6 vs DipNet SL | average score: 15.5% -> 20.4% | Improved |
| train network to imitate time-average of BRs (FPPI-1) instead of averaging checkpoints explicitly (FPPI-2) (ablation) | BRPI · No-Press Diplomacy against final BRPI agents | winrate | Hurt |
| stochastic SBR instead of exact best response in iterated best response (ablation) | IBR · Blotto | convergence | Improved |
| train RL agents with BRPI | Diplomacy agents · few-shot exploitability | exploitability | Mixed |
Closest papers by meaning
- No-Press Diplomacy from Scratch (NeurIPS 2021 · similarity 0.93)
- No-Press Diplomacy: Modeling Multi-Agent Gameplay (NeurIPS 2019 · similarity 0.91)
- Mastering the Game of No-Press Diplomacy via Human-Regularized Reinforcement Learning and Planning (ICLR 2023 · similarity 0.91)
- Explicit Exploration for High-Welfare Equilibria in Game-Theoretic Multiagent Reinforcement Learning (ICML 2025 · similarity 0.89)
- DipLLM: Fine-Tuning LLM for Strategic Decision-making in Diplomacy (ICML 2025 · similarity 0.87)
- Solving Continuous Mean Field Games: Deep Reinforcement Learning for Non-Stationary Dynamics (NeurIPS 2025 · 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.
