伪造和今后的业绩
Falsification and future performance
摘要：
我们从信息论的角度重新定义了统计学习理论中的两种能力度量：经验VC熵和经验拉德马赫复杂度。我们表明，这些能力的措施计数的数量的假设，一个学习算法伪造的数据集，当它发现分类器在其剧目最小化的经验风险。因此，预测器在未知数据上的未来表现部分取决于学习者伪造了多少假设。作为一个推论，我们表明，经验VC熵量化的消息长度的真实假设在一个特定的概率分布，所谓的实际剧目的最佳代码。
英文摘要：
We information-theoretically reformulate two measures of capacity from statistical learning theory: empirical VC-entropy and empirical Rademacher complexity. We show these capacity measures count the number of hypotheses about a dataset that a learning algorithm falsifies when it finds the classifier in its repertoire minimizing empirical risk. It then follows from that the future performance of predictors on unseen data is controlled in part by how many hypotheses the learner falsifies. As a corollary we show that empirical VC-entropy quantifies the message length of the true hypothesis in the optimal code of a particular probability distribution, the so-called actual repertoire.
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。
