On Loss Functions for Deep Neural Networks in Classification读后感

最新推荐文章于 2024-03-15 18:38:39 发布

原创最新推荐文章于 2024-03-15 18:38:39 发布 · 419 阅读

0 ·

CC 4.0 BY-SA版权

机器学习专栏收录该内容

12 篇文章

订阅专栏

本文探讨了在分类问题中，不同损失函数的表现与适用场景。对于追求准确率的研究，平方合页损失（Squared Hinge Loss）因其更快的收敛速度和更好的性能成为优选。它在训练集标签噪声和输入空间噪声方面更为鲁棒。然而，在高度嘈杂的数据集上，论文中详述的期望损失则表现最佳。此外，非经典损失函数如Tanimoto损失和Cauchy-Schwarz发散值得进一步研究。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

分类问题中的另一类loss函数

在这里插入图片描述

In particular, for purely accuracy focused research, squared hinge loss seems
to be a better choice at it converges faster as well as provides better performance.
It is also more robust to noise in the training set labelling and
slightly more robust to noise in the input space. However, if one works
with highly noised dataset (both input and output spaces) – the expectation
losses described in detail in this paper – seem to be the best choice,
both from theoretical and empirical perspective.
At the same time this topic is far from being exhausted, with a large
amount of possible paths to follow and questions to be answered. In
particular, non-classical loss functions such as Tanimoto loss(噪声大的情况) and CauchySchwarz
Divergence(无噪声) are worth further investigation.