Research map / Learning theory and algorithms / Kernels and regression theory
Optimal Regularization for Performative Learning
Edwige Cyffers, Alireza Mirrokni, Marco Mondelli
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 |
|---|---|---|---|
| increase ridge regularization to the optimum as performative strength b-bar grows (n=4000) | ridge regression under repeated retraining · Housing and LSAC (n=4000) | optimal regularizer and excess risk | Improved |
| stronger performative effect with few samples (n=100) | optimally regularized ridge regression · LSAC (n=100) | excess risk | Improved |
| add performative effect on spurious features (c) | ridge regression · Housing and LSAC | excess risk | No clear effect |
| replace ridge with dropout regularization (ablation) | performative regression · Housing and LSAC | excess risk | Mixed |
| replace ridge with Lasso regularization (ablation) | performative regression · Housing and LSAC | optimal regularizer | Mixed |
| replace ridge with elastic net (equal l1 and l2) (ablation) | performative regression · Housing and LSAC | excess risk | No clear effect |
| add l2 regularization to a neural network under strategic shift | neural network classifier · GiveMeSomeCredit | final test accuracy | Improved |
Closest papers by meaning
- Optimal Classification under Performative Distribution Shift (NeurIPS 2024 · similarity 0.90)
- Plug-in Performative Optimization (ICML 2024 · similarity 0.88)
- Optimal Ridge Regularization for Out-of-Distribution Prediction (ICML 2024 · similarity 0.88)
- On the Mechanisms of Weak-to-Strong Generalization: A Theoretical Perspective (NeurIPS 2025 · similarity 0.86)
- Performative Prediction with Bandit Feedback: Learning through Reparameterization (ICML 2024 · similarity 0.86)
- On the Impact of Performative Risk Minimization for Binary Random Variables (ICML 2025 · similarity 0.86)
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.
