arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 1206.4663cs.LGstat.ML
收藏

复合多类损耗的凸性与设计

The Convexity and Design of Composite Multiclass Losses

Research School of Computer Science, The Australian National University(澳大利亚国立大学计算机科学研究院);NICTA;Tsinghua University(清华大学)

Mark Reid, Robert Williamson, Peng Sun

摘要:

我们考虑用于多类预测的复合损失函数,其包括适当的(即,费雪一致性)损失的概率分布和逆链接函数。我们建立条件,他们的(强)凸性,并探讨其影响。我们还展示了如何分离的关注,通过使用这种复合表示允许设计的家庭损失具有相同的贝叶斯风险。

英文摘要:

We consider composite loss functions for multiclass prediction comprising a proper (i.e., Fisher-consistent) loss over probability distributions and an inverse link function. We establish conditions for their (strong) convexity and explore the implications. We also show how the separation of concerns afforded by using this composite representation allows for the design of families of losses with the same Bayes risk.

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。

↑