应对智能体策略性行为，实现稳定可信的预测置信度量化。
Strategic Conformal Prediction
- 引入策略性置信预测框架，考虑预测影响环境的动态反馈。
- 理论保证覆盖率达95%以上，且对模型误设具鲁棒性。
- 适合部署于博弈场景、自适应系统等需防策略干扰的领域。
当机器学习模型投入使用时，其预测结果会改变环境，使更理性的参与者根据预测调整行为以谋求自身利益。在此背景下，现有不确定性量化方法失效。本文提出一种新框架——策略性置信预测（Strategic Conformal Prediction），可在该动态环境中实现稳健的不确定性量化。该框架具备一系列分布无关的理论保障，涵盖边际覆盖率、训练条件覆盖率、紧致性及对模型误设的鲁棒性。实验验证表明，面对任意策略性扰动，本方法表现优异，而其他方法则迅速失效。
原文摘要 · Abstract (English)
When a machine learning model is deployed, its predictions can alter its environment, as better informed agents strategize to suit their own interests. With such alterations in mind, existing approaches to uncertainty quantification break. In this work we propose a new framework, Strategic Conformal Prediction, which is capable of robust uncertainty quantification in such a setting. Strategic Conformal Prediction is backed by a series of theoretical guarantees spanning marginal coverage, training-conditional coverage, tightness and robustness to misspecification that hold in a distribution-free manner. Experimental analysis further validates our method, showing its remarkable effectiveness in face of arbitrary strategic alterations, whereas other methods break.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。