arXiv：cs.LG· Giovani Valdrighi, Isabel Valera, Marcos Medeiros Raimundo·· 4 小时前AI 评分39
选择性标签下的表演性预测
Performative Prediction with Selective Labels
AI 导读
该研究形式化了带选择性标签的表演性预测问题：由于只能观测到被接受人群的标签，仅基于已观测数据重训练会误导重复风险最小化（RRM），破坏其收敛到稳定模型的保证。作者提出基于正标签概率置信区间的鲁棒优化目标，使 RRM 保持在距真实稳定点有界距离内，并在条件标签分布的敏感性假设下利用历史已接受数据收紧置信区间。在带公平正则化的借贷实验中，该方法性能接近可获取完整标签的 RRM。
正文
Abstract:Many social applications of machine learning exhibit performative effects: population behavior changes in response to deployed models. Performative prediction studies this interaction through a distribution map that relates each model to the population distribution it induces. One of the main results in this framework showed that repeated risk minimization (RRM), which updates models by retraining on the most recent data, can converge to a stable model that minimizes risk on its own induced distribution. However, existing analyses typically assume access to the complete distributions of features and labels after model deployment, ignoring the possibility of selective labels: observing labels only for the accepted subset of the population. In this work, we formalize performative prediction with selective labels and show that retraining only on observed data can misguide the retraining procedure and undermine the guarantees of convergence to a stable solution. We then propose a worst-case objective based on knowledge of a confidence interval on the probability of a positive label. Applying RRM to this objective permits us to remain within a bounded distance to the true stable point. Under a sensitivity assumption on the conditional label distribution, we further show how previously accepted data can tighten these confidence intervals over time. Experiments in a lending application with fairness regularization show that our robust optimization approach closely matches the performance of RRM with complete label access.
| Comments: | Accepted at NeurIPS 2026. Camera-ready version |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.08272 [cs.LG] |
| (or arXiv:2610.08272v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08272 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Giovani Valdrighi [view email]
[v1]
Tue, 6 Oct 2026 12:45:43 UTC (738 KB)
来源：arXiv：cs.LG · arxiv.org