2024/08/16 by Anmol Kabra, Kumar Kshitij Patel, Kabra, Anmol +1
Social Sciences · #Computer Science and Game Theory (cs.GT) #Educational Tools and Methods #FOS: Computer and information sciences #Machine Learning (cs.LG)
paper · pdf · doi:10.48550/arxiv.2408.08499
openalex publication_date 2024/08/16 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
We study stochastic optimization in the context of performative shifts, where the data distribution changes in response to the deployed model. We demonstrate that naive retraining can be provably suboptimal even for simple distribution shifts. The issue worsens when models are retrained given a finite number of samples at each retraining step. We show that adding regularization to retraining corrects both of these issues, attaining provably optimal models in the face of distribution shifts. Our work advocates rethinking how machine learning models are retrained in the presence of performative effects.