arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.04385cs.LGcs.ITmath.ITmath.STstat.TH
收藏

基于相对熵正则化的经验风险最小化:最优性和灵敏度分析

Empirical Risk Minimization with Relative Entropy Regularization: Optimality and Sensitivity Analysis

INRIA(法国国家信息与自动化研究所);Université de la Polynésie Française(法属波利尼西亚大学);University of Sheffield(谢菲尔德大学);National Chao Tung University(国立交通大学);Princeton University(普林斯顿大学)

Samir M. Perlaza, Gaetan Bisson, Iñaki Esnaola, Alain Jean-Marie, Stefano Rini

摘要:

研究了相对熵正则化的经验风险最小化问题(ERM-RER)的最优性和灵敏度,其中参考测度是σ-有限测度而不是概率测度.这种泛化允许在模型集合上结合先验知识时具有更大程度的灵活性。在这种情况下,正则化参数,参考措施,风险函数,和ERM-RER问题的解决方案引起的经验风险的相互作用的特点。这个特征产生了正则化参数存在的必要和充分条件,该正则化参数以任意高的概率实现任意小的经验风险。研究了ERM-RER问题的期望经验风险对偏离解的敏感性。然后,敏感性被用来提供预期的经验风险的上限和下限。此外,它示出的期望的灵敏度是上界,直到一个常数因子,由模型和数据集之间的lautum信息的平方根。

英文摘要:

The optimality and sensitivity of the empirical risk minimization problem with relative entropy regularization (ERM-RER) are investigated for the case in which the reference is a sigma-finite measure instead of a probability measure. This generalization allows for a larger degree of flexibility in the incorporation of prior knowledge over the set of models. In this setting, the interplay of the regularization parameter, the reference measure, the risk function, and the empirical risk induced by the solution of the ERM-RER problem is characterized. This characterization yields necessary and sufficient conditions for the existence of a regularization parameter that achieves an arbitrarily small empirical risk with arbitrarily high probability. The sensitivity of the expected empirical risk to deviations from the solution of the ERM-RER problem is studied. The sensitivity is then used to provide upper and lower bounds on the expected empirical risk. Moreover, it is shown that the expectation of the sensitivity is upper bounded, up to a constant factor, by the square root of the lautum information between the models and the datasets.

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。

↑