Research map / Multi-agent systems and games / Games and equilibria
Learning Mean-Field Games
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 |
|---|---|---|---|
| increase inner iterations from 1000 to 10000 (ablation) | GMF-Q · Repeated auction GMFG | Delta Q: 0.21263 -> 0.0989 | Helped |
| use GMF-Q instead of IL for N=20 players | N-player repeated auction · N = 20, |S|=|A|=10 | error C(pi): 0.220 -> 0.065 | Helped |
| use GMF-Q instead of MF-Q for N=20 players | N-player repeated auction · N = 20, |S|=|A|=10 | error C(pi): 0.101 -> 0.065 | Helped |
| use Boltzmann-policy Q-learning (GMF-Q) instead of the naive fixed-point Q-learning (ablation) | mean-field Q-learning · Repeated auction GMFG | convergence | Helped |
| increase number of states from 10 to 30 and 100 (ablation) | GMF-Q · Repeated auction GMFG | convergence of population distribution | No clear effect |
| increase number of players N while also increasing state/action space size | GMF-Q vs IL and MF-Q · N-player repeated auction | error C(pi) | Mixed |
Closest papers by meaning
- Learning Discrete-Time Major-Minor Mean Field Games (AAAI 2024 · similarity 0.89)
- Stochastic Semi-Gradient Descent for Learning Mean Field Games with Population-Aware Function Approximation (ICLR 2025 · similarity 0.89)
- Bounded Rationality Equilibrium Learning in Mean Field Games (AAAI 2025 · similarity 0.88)
- Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean Field Games (ICML 2025 · similarity 0.88)
- Bench-MFG: A Benchmark Suite for Learning in Stationary Mean Field Games (NeurIPS 2026 · similarity 0.88)
- 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.