A Survey of Learning Criteria Going Beyond the Usual Risk

A Survey of Learning Criteria Going Beyond the Usual Risk
复制标题

超越通常风险的学习标准调查

DOI:
10.1613/jair.1.15000
复制
发表时间:
2021
影响因子:
1.8
通讯作者:
Kazuki Tanabe
Kazuki Tanabe
中科院分区:
数学3区
文献类型:
--
作者:
Matthew J. Holland;Kazuki Tanabe

文献摘要

参考文献

被引文献

相似文献

实际上,所有机器学习任务都通过某种形式的损失函数来表征,并且“良好性能”通常是根据在测试数据的随机抽取上取得足够小的平均损失来衡量的。虽然平均性能优化在直觉上合理,在理论上便于分析,在实践中易于实施,但这样的选择会带来权衡。在这项工作中,我们综述并介绍了用于设计和评估机器学习算法的多种非传统标准,将经典范式置于恰当的历史背景中,并提出了一种学习问题的观点,该观点强调“是什么构成了理想的损失分布?”这一问题,而非默认使用期望损失。
Virtually all machine learning tasks are characterized using some form of loss function, and “good performance” is typically stated in terms of a sufficiently small average loss, taken over the random draw of test data. While optimizing for performance on average is intuitive, convenient to analyze in theory, and easy to implement in practice, such a choice brings about trade-offs. In this work, we survey and introduce a wide variety of non-traditional criteria used to design and evaluate machine learning algorithms, place the classical paradigm within the proper historical context, and propose a view of learning problems which emphasizes the question of “what makes for a desirable loss distribution?” in place of tacit use of the expected loss.
DOI: --
发表时间: 2021
期刊: Advances in neural information processing systems
影响因子: --
作者:
Finocchiaro, Jessica;Frongillo, Rafael;Waggoner, Bo
通讯作者: Waggoner, Bo
DOI: --
发表时间: 2018-10
期刊: --
影响因子: --
作者:
S. Samadi;U. Tantipongpipat;Jamie Morgenstern;Mohit Singh;S. Vempala
通讯作者: S. Samadi;U. Tantipongpipat;Jamie Morgenstern;Mohit Singh;S. Vempala
DOI: --
发表时间: 2021-09
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Tian Li;Ahmad Beirami;Maziar Sanjabi;Virginia Smith
通讯作者: Tian Li;Ahmad Beirami;Maziar Sanjabi;Virginia Smith