基于选择性分类的标签噪声转移矩阵估计及其性能保证
Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification
浏览论文内容
中文总结 AI 辅助
针对现有标签噪声转移矩阵估计依赖脆弱类后验且缺乏有限样本保证的问题，提出基于单侧选择性分类的新方法，绕过类后验估计并提供性能保证与有效算法。
中文摘要 AI 辅助
现代机器学习高度依赖大规模数据集，但在大规模下获取高质量标注往往代价高昂。因此，从带噪标签的数据中学习已变得普遍，这使得准确估计标签噪声转移矩阵变得至关重要。然而，现有的转移矩阵估计器依赖于脆弱的类后验估计，并且不提供有限样本性能保证。在这项工作中，我们提出了一种基于单侧选择性分类来估计转移矩阵的新方法。该方法绕过了类后验估计，提供了有限样本性能保证，并利用了灵活的二元分类学习方法。此外，我们引入了有效的算法来实现所提出的方法，并提供了其精细的有限样本性能界限。
英文摘要
Modern machine learning depends heavily on massive datasets, but obtaining high-quality annotations at scale is often expensive. As a result, learning from noisily-labeled data has become common, making accurate estimation of the label-noise transition matrix crucial. However, existing transition matrix estimators rely on the fragile estimation of class-posteriors and do not provide finite-sample performance guarantees. In this work, we propose a novel methodology to estimate the transition matrix based on one-sided selective classification. This approach bypasses class-posterior estimation, provides finite-sample performance guarantees, and leverages flexible learning methods for binary classification. Moreover, we introduce effective algorithms to implement the proposed methodology and provide their refined finite-sample performance bounds.
发表机构
- Basque Center of Applied Mathematics (BCAM)（巴斯克应用数学中心（BCAM））
- IKERBASQUE-Basque Foundation for Science（IKERBASQUE-巴斯克科学基金会）
- University of Michigan（密歇根大学）
机构由 AI 辅助整理，请以论文原文为准。
