arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2210.03069cs.LG
收藏

PathProx:一种用于权重衰减正则化深度神经网络的近端梯度算法

PathProx: A Proximal Gradient Algorithm for Weight Decay Regularized Deep Neural Networks

University of Wisconsin, Madison(威斯康星大学麦迪逊分校)

Liu Yang, Jifan Zhang, Joseph Shenouda, Dimitris Papailiopoulos, Kangwook Lee, Robert D. Nowak

AI总结:

本文提出 PathProx 近端梯度算法,利用 ReLU 网络权重衰减目标的等价重构形式,实现比 SGD 更快的稀疏解收敛。

摘要:

权重衰减是深度学习中使用最广泛的正则化形式之一,已被证明可以改善泛化能力和鲁棒性。驱动权重衰减的优化目标是一个损失之和加上一个与权重平方和成比例的项。本文认为,对于该目标,随机梯度下降(SGD)可能是一种低效的算法。对于具有 ReLU 激活的神经网络,权重衰减目标的解等价于另一个目标的解,在该目标中,正则化项变为与每个 ReLU 神经元相关联的输入和输出权重的 $\ell_2$(非平方)范数乘积之和。这种替代的(且实际上等价的)正则化提出了一种用于网络训练的新型近端梯度算法。理论和实验支持这种新的训练方法,表明它能够更快地收敛到与标准权重衰减训练共享的稀疏解。

英文摘要:

Weight decay is one of the most widely used forms of regularization in deep learning, and has been shown to improve generalization and robustness. The optimization objective driving weight decay is a sum of losses plus a term proportional to the sum of squared weights. This paper argues that stochastic gradient descent (SGD) may be an inefficient algorithm for this objective. For neural networks with ReLU activations, solutions to the weight decay objective are equivalent to those of a different objective in which the regularization term is instead a sum of products of $\ell_2$ (not squared) norms of the input and output weights associated with each ReLU neuron. This alternative (and effectively equivalent) regularization suggests a novel proximal gradient algorithm for network training. Theory and experiments support the new training approach, showing that it can converge much faster to the sparse solutions it shares with standard weight decay training.

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。

↑