基于Hamming查询序列重构的最优多类过拟合
Optimal multiclass overfitting by sequence reconstruction from Hamming queries
摘要：
在机器学习中过度重用测试数据集的主要问题是它可能导致过拟合。多类分类最近被证明比二进制分类更能抵抗过拟合。在COLT 2019的一个开放问题中，Feldman、Frostig和Hardt要求描述过拟合偏差量与类数$m$、准确性查询数$k$和数据集$n$中示例数的依赖性。我们解决这个问题，并确定在多类分类可能的过拟合量。我们提供了计算效率高的算法，实现过拟合偏差的$\tildeΘ(\max\{\sqrt{{k}/{(mn)}}, k/n\})$，匹配已知的上限。
英文摘要：
A primary concern of excessive reuse of test datasets in machine learning is that it can lead to overfitting. Multiclass classification was recently shown to be more resistant to overfitting than binary classification. In an open problem of COLT 2019, Feldman, Frostig, and Hardt ask to characterize the dependence of the amount of overfitting bias with the number of classes $m$, the number of accuracy queries $k$, and the number of examples in the dataset $n$. We resolve this problem and determine the amount of overfitting possible in multi-class classification. We provide computationally efficient algorithms that achieve overfitting bias of $\tildeΘ(\max\{\sqrt{{k}/{(mn)}}, k/n\})$, matching the known upper bounds.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
