摘要:
通常而言,损失函数由损失项(loss term)和正则项(regularization term)组成。发现一份不错的介绍资料: http://www.ics.uci.edu/~dramanan/teaching/ics273a_winter08/lectures/lecture14.pdf (题名 阅读全文
摘要:
1. An overview of gradient descent optimization algorithms http://ruder.io/optimizing-gradient-descent/index.html https://arxiv.org/abs/1609.04747 2. 阅读全文
摘要:
http://blog.csdn.net/zouxy09/article/details/24971995/ L1正则化及其推导 http://blog.csdn.net/zouxy09/article/details/24971995/ L1正则化及其推导 Laplace(拉普拉斯)先验与L1正则 阅读全文