Research map / Optimization / Nonconvex and smooth optimization
Stochastic Gradient Descent-Ascent and Consensus Optimization for Smooth Games: Convergence Analysis under Expected Co-coercivity
Nicolas Loizou, Hugo Berard, Gauthier Gidel, Ioannis Mitliagkas, Simon Lacoste-Julien
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 |
|---|---|---|---|
| Combine SGDA and SHGD updates with positive weights | SCO · Strongly-monotone quadratic games | relative distance to optimality | Mixed |
| Use a decreasing step-size after an initial constant-step phase | SGDA · Strongly-monotone quadratic games | precision | Improved |
| Use a decreasing step-size after an initial constant-step phase | SCO · Strongly-monotone quadratic games | precision | Improved |
| Increase the game condition number | SGDA · Strongly-monotone quadratic games | convergence speed | Hurt |
| Increase the game condition number | SHGD and SCO · Strongly-monotone quadratic games | convergence speed | No clear effect |
| Remove the SGDA component by setting its step size to zero (ablation) | SCO · Quasi-strongly monotone variational inequalities | convergence rate | Hurt |
| Remove the SGDA component by setting its step size to zero (ablation) | SCO · Variationally stable problems with mu=0 | convergence neighborhood | Improved |
| Remove the Hamiltonian-gradient component by setting its step size to zero (ablation) | SCO · Quasi-strongly monotone variational inequalities | convergence rate | Hurt |
Closest papers by meaning
- Stochastic Hamiltonian Gradient Methods for Smooth Games (ICML 2020 · similarity 0.90)
- Decoupled SGDA for Games with Intermittent Strategy Communication (ICML 2025 · similarity 0.87)
- Consensus Based Stochastic Optimal Control (ICML 2025 · similarity 0.87)
- Towards Runtime Analysis of Population-Based Co-evolutionary Algorithms on Sparse Binary Zero-Sum Game (AAAI 2025 · similarity 0.87)
- Stability and Generalization of Stochastic Compositional Gradient Descent Algorithms (ICML 2024 · similarity 0.87)
- Gradient-Variation Online Learning under Generalized Smoothness (NeurIPS 2024 · 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.
