Research map / Multi-agent systems and games / Games and equilibria
Asynchronous Gradient Play in Zero-Sum Multi-agent 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 |
|---|---|---|---|
| Add two-timescale, delay-aware learning-rate separation | OMWU · Zero-sum polymatrix games with constant known delays; ε-NE | iteration complexity: \widetilde{\mathcal{O}}\big({d_\mathsf{max}^2\norm{A}_\infty^2(\gamma+1)^2}/{\epsilon^2}\big) -> \widetilde \mathcal{O}\ | Helped |
| Add delayed feedback | Single-timescale OMWU · Zero-sum polymatrix games with random statistical delays; ε-NE | iteration complexity: \widetilde{\mathcal{O}}(\epsilon^{-1}) -> \widetilde{\mathcal{O}}(\epsilon^{-2}) | Hurt |
| Use two-timescale OMWU rather than single-timescale OMWU | OMWU · Randomly generated zero-sum entropy-regularized polymatrix games with permuted bounded delay | convergence rate to QRE | Helped |
| Set the extrapolation learning rate to the theory-suggested delay-aware value (ablation) | Two-timescale OMWU · Randomly generated zero-sum entropy-regularized polymatrix games with bounded delay | convergence performance | Helped |
Closest papers by meaning
- Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms (NeurIPS 2024 · similarity 0.91)
- Optimistic Policy Gradient in Multi-Player Markov Games with a Single Controller: Convergence beyond the Minty Property (AAAI 2024 · similarity 0.91)
- When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation (NeurIPS 2026 · similarity 0.89)
- Learning from Delayed Feedback in Games via Extra Prediction (NeurIPS 2025 · similarity 0.89)
- From Average-Iterate to Last-Iterate Convergence in Games: A Reduction and Its Applications (NeurIPS 2025 · similarity 0.89)
- Solving Zero-Sum Convex Markov Games (ICML 2025 · similarity 0.89)
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.