arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1003.0516cs.LG
收藏

基于损失秩原理的模型选择

Model Selection with the Loss Rank Principle

NICTA(澳大利亚国家信息与通信技术研究院);National University of Singapore(新加坡国立大学)

Marcus Hutter, Minh-Ngoc Tran

摘要:

统计学和机器学习中的一个关键问题是自动选择“正确的”模型复杂度,例如,在k最近邻(kNN)回归中要平均的邻居的数量或在多项式回归中的多项式次数。我们提出了一个新的原则-损失秩原则(LoRP)-在回归和分类的模型选择。它基于损失等级,计算有多少其他(虚构的)数据会更好地拟合。LoRP选择具有最小损失秩的模型。与大多数惩罚最大似然变量(AIC,BIC,MDL)不同,LoRP仅取决于回归函数和损失函数。它在没有随机噪声模型的情况下工作,并且直接适用于任何非参数回归量,如kNN。

英文摘要:

A key issue in statistics and machine learning is to automatically select the "right" model complexity, e.g., the number of neighbors to be averaged over in k nearest neighbor (kNN) regression or the polynomial degree in regression with polynomials. We suggest a novel principle - the Loss Rank Principle (LoRP) - for model selection in regression and classification. It is based on the loss rank, which counts how many other (fictitious) data would be fitted better. LoRP selects the model that has minimal loss rank. Unlike most penalized maximum likelihood variants (AIC, BIC, MDL), LoRP depends only on the regression functions and the loss function. It works without a stochastic noise model, and is directly applicable to any non-parametric regressor, like kNN.

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。

↑