arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1709.01427stat.MLcs.LGcs.NE
收藏

随机梯度下降:尽可能快,但不要更快

Stochastic Gradient Descent: Going As Fast As Possible But Not Faster

UCSF(加州大学旧金山分校);INRIA-CNRS-UPSud-UPSay(法国国家信息与自动化研究所-法国国家科学研究中心-巴黎南大学-巴黎萨克雷大学);TAU, U. Paris-Sud(计算机科学研究实验室,巴黎南大学);CNRS-UPSud-INRIA-UPSay(法国国家科学研究中心-巴黎南大学-法国国家信息与自动化研究所-巴黎萨克雷大学)

Alice Schoenauer-Sebag, Marc Schoenauer, Michèle Sebag

摘要:

当应用于训练深度神经网络时,随机梯度下降(SGD)通常会导致稳定的进展阶段,被损失和梯度范数爆炸的灾难性事件中断。缓解此类事件的一种可能方法是减缓学习过程。本文提出了一种新的方法来控制SGD学习率,使用两个统计测试。第一个,旨在快速学习,比较归一化梯度向量的动量和随机单位向量的动量,并相应地优雅地增加或减少学习速率。第二个是变化点检测测试,旨在检测灾难性的学习事件;一旦触发,学习率立即减半。加速和减慢学习速率的能力都允许所提出的称为SALeRA的方法尽可能快地学习,但不会更快。在标准基准上的实验表明,SALeRA在实践中表现良好,并与最先进的相比。

英文摘要:

When applied to training deep neural networks, stochastic gradient descent (SGD) often incurs steady progression phases, interrupted by catastrophic episodes in which loss and gradient norm explode. A possible mitigation of such events is to slow down the learning process. This paper presents a novel approach to control the SGD learning rate, that uses two statistical tests. The first one, aimed at fast learning, compares the momentum of the normalized gradient vectors to that of random unit vectors and accordingly gracefully increases or decreases the learning rate. The second one is a change point detection test, aimed at the detection of catastrophic learning episodes; upon its triggering the learning rate is instantly halved. Both abilities of speeding up and slowing down the learning rate allows the proposed approach, called SALeRA, to learn as fast as possible but not faster. Experiments on standard benchmarks show that SALeRA performs well in practice, and compares favorably to the state of the art.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑