...
首页> 外文期刊>Journal of the European Mathematical Society: JEMS >Risk minimization by median-of-means tournaments
【24h】

Risk minimization by median-of-means tournaments

机译:由中位数锦标赛的风险最小化

获取原文
获取原文并翻译 | 示例
           

摘要

We consider the classical statistical learning/regression problem, when the value of a real random variable Y is to be predicted based on the observation of another random variable X. Given a class of functions F and a sample of independent copies of (X, Y), one needs to choose a function (f) over cap from F such that b (f) over cap (X) approximates Y as well as possible, in the mean-squared sense. We introduce a new procedure, the so-called median-of-means tournament, that achieves the optimal tradeoff between accuracy and confidence under minimal assumptions, and in particular outperforms classical methods based on empirical risk minimization.
机译:我们考虑古典统计学习/回归问题,当基于另一个随机变量X的观察来预测真实随机变量Y的值。给定一类功能f和(x,y的独立副本 ),需要选择从F的帽上的功能(f),使得在帽(x)上方的B(f)近似于y,在平均方形的意义上。 我们介绍了一种新的程序,所谓的中音锦标赛,实现了在最小假设下的准确性和信心之间的最佳权衡,特别优于基于经验风险最小化的经典方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号