Research map / Optimization / Optimizers
Stochastic Optimization for Performative Prediction
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 |
|---|---|---|---|
| use greedy deploy instead of lazy deploy when performative effects are weak | stochastic optimization for performative prediction · synthetic Gaussian | convergence to performative stability | Helped |
| use lazy deploy instead of greedy deploy when performative effects are strong | stochastic optimization for performative prediction · synthetic Gaussian | convergence to performative stability | Helped |
| use a shorter lazy deployment schedule with smaller α (ablation) | lazy deploy · synthetic Gaussian | initial optimization progress | Helped |
| use a longer lazy deployment schedule with larger α (ablation) | lazy deploy · synthetic Gaussian | long-run progress | Helped |
| use lazy deploy instead of greedy deploy | stochastic optimization for performative prediction · strategic classification, ε=100 | convergence to performative stability | Helped |
| increase α in the lazy deployment schedule (ablation) | lazy deploy · strategic classification, ε=100 | convergence to performative stability | Helped |
| use greedy deploy instead of lazy deploy | stochastic optimization for performative prediction · strategic classification, ε=10^-3 | convergence to performative stability | Helped |
Closest papers by meaning
- Stochastic Optimization Schemes for Performative Prediction with Nonconvex Loss (NeurIPS 2024 · similarity 0.91)
- Plug-in Performative Optimization (ICML 2024 · similarity 0.91)
- Two-timescale Derivative Free Optimization for Performative Prediction with Markovian Data (ICML 2024 · similarity 0.89)
- Performative Prediction with Bandit Feedback: Learning through Reparameterization (ICML 2024 · similarity 0.89)
- The Stability of Online Algorithms in Performative Prediction (NeurIPS 2026 · similarity 0.89)
- Optimal Classification under Performative Distribution Shift (NeurIPS 2024 · 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.