仅用一次前向传播，即可实现鲁棒的置信预测
One Sample is Enough to Make Conformal Prediction Robust
- 用单次随机扰动输入，认证预测过程本身以获得鲁棒性
- 相比现有方法（如100次前向传播），平均预测集更小
- 适用于分类与回归，对模型类型无要求
对于任意黑盒模型，共形预测（CP）可生成包含真实标签的高概率预测集。鲁棒共形预测（RCP）将这一保证扩展至预定义范围内的最坏情况噪声。现有主流方法采用随机平滑，虽适用于所有黑盒模型且预测集更小，但需对每个输入进行多次前向传播，计算成本高。本文发现，即使仅使用一次前向传播于随机扰动输入，共形预测也能获得一定鲁棒性。我们提出单样本鲁棒共形预测（RCP1），利用任意二值校准函数，在无需多次推理的情况下生成更小的鲁棒预测集。核心思想是认证共形过程而非单个符合度分数。该方法适用于分类与回归任务，且与具体模型无关。进一步地，我们将此思路拓展至基于平滑的鲁棒共形风险控制。
原文摘要 · Abstract (English)
For any black-box model, conformal prediction (CP) returns prediction sets guaranteed to include the true label with high adjustable probability. Robust CP (RCP) extends the guarantee to the worst case noise up to a pre-defined magnitude. For RCP, a well-established approach is to use randomized smoothing since it is applicable to any black-box model and provides smaller sets compared to deterministic methods. However, smoothing-based robustness requires many model forward passes per each input which is computationally expensive. We show that conformal prediction attains some robustness even with a single forward pass on a randomly perturbed input. Using any binary certificate we propose a single sample robust CP (RCP1). Our approach returns robust sets with smaller average set size compared to SOTA methods which use many (e.g. 100) passes per input. Our key insight is to certify the conformal procedure itself rather than individual conformity scores. Our approach is agnostic to the task (classification and regression). We further extend our approach to smoothing-based robust conformal risk control.
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。