Research map / Learning theory and algorithms / Submodular and game algorithms
Chaos, Extremism and Optimism: Volume Analysis of Learning in Games
Yun Kuen Cheung, Georgios Piliouras
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 |
|---|---|---|---|
| Use MWU in a zero-sum game rather than OMWU | MWU and OMWU · Zero-sum games | dual-space volume change: MWU expands volume; OMWU contracts volume | Mixed |
| Use OMWU in a coordination game rather than MWU | MWU and OMWU · Coordination games | dual-space volume change: OMWU expands volume; MWU contracts volume | Mixed |
| Apply OMWU to zero-sum games | OMWU · Zero-sum games | dual-space volume change: volume contracts exponentially | Improved |
| Apply OMWU to non-trivial coordination games | OMWU · Coordination games | dual-space volume change: (1 + ε²δ²(α₁)²/4)^T | Hurt |
| Use MWU in zero-sum games | MWU · Zero-sum games | avoidance of bad states | Hurt |
| Use MWU in zero-sum games satisfying the theorem conditions | MWU · Zero-sum games | reaching extremal domain: threshold δ | Hurt |
| Use MWU in Rock-Paper-Scissors | MWU · Rock-Paper-Scissors | extremism periods: 22 periods within 30000 time steps; each period has length around 350 | Hurt |
Closest papers by meaning
- When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation (NeurIPS 2026 · similarity 0.84)
- Fast Last-Iterate Convergence of Learning in Games Requires Forgetful Algorithms (NeurIPS 2024 · similarity 0.84)
- A Geometric Decomposition of Finite Games: Convergence vs. Recurrence under Exponential Weights (ICML 2024 · similarity 0.82)
- On the Optimality of Dilated Entropy and Lower Bounds for Online Learning in Extensive-Form Games (NeurIPS 2024 · similarity 0.82)
- Explaining Probabilistic Models with Distributional Values (ICML 2024 · similarity 0.81)
- Prediction Accuracy of Learning in Games : Follow-the-Regularized-Leader meets Heisenberg (ICML 2024 · similarity 0.81)
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.
