2020年2月16日
Accelerating Deep Learning by Focusing on the Biggest Losers
摘要： Accelerating Deep Learning by Focusing on the Biggest Losers 概 思想很简单, 在训练网络的时候, 每个样本都会产生一个损失$\mathcal{L}(f(x_i),y_i)\(, 训练的模式往往是批训练, 将一个批次\)\sum_i \ma 阅读全文
posted @ 2020-02-16 21:24 馒头and花卷 阅读(383) 评论(0) 推荐(0)
Not All Samples Are Created Equal: Deep Learning with Importance Sampling
摘要： Katharopoulos A, Fleuret F. Not All Samples Are Created Equal: Deep Learning with Importance Sampling[J]. arXiv: Learning, 2018. @article{katharopoulo 阅读全文
posted @ 2020-02-16 20:42 馒头and花卷 阅读(740) 评论(2) 推荐(0)