The maths that drives AI
成果类型：
Editorial Material
署名作者：
Brooks, Michael
刊物名称：
Nature
ISSN/ISSBN：
0028-5902
DOI：
10.1038/d41586-024-02185-z
发表日期：
2024-07-04
页码：
244-246
关键词：
摘要：
Loss functions measure algorithmic errors in artificial-intelligence models, but there's more than one way to do that. Here's why the right function is so important.