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 1604.06985cs.LG
收藏

使用特征值衰减正则化器的深度学习

Deep Learning with Eigenvalue Decay Regularizer

KU Leuven(鲁汶大学)

Oswaldo Ludwig

摘要:

本文扩展了我们之前使用特征值衰减正则化神经网络的工作,通过采用主特征值的软近似,以便计算其与突触权重相关的导数,从而应用反向传播,这是深度学习的主要需求。此外,我们将之前的理论分析扩展到深度神经网络和多类分类问题。我们的方法在Keras中实现为一个额外的正则化器,Keras是一个用Python编写的模块化神经网络库,并在基准数据集Reuters Newswire Topics Classification,用于二进制情感分类的IMDB数据库,手写数字的MNIST数据库和用于图像分类的CIFAR-10数据集中进行评估。

英文摘要:

This paper extends our previous work on regularization of neural networks using Eigenvalue Decay by employing a soft approximation of the dominant eigenvalue in order to enable the calculation of its derivatives in relation to the synaptic weights, and therefore the application of back-propagation, which is a primary demand for deep learning. Moreover, we extend our previous theoretical analysis to deep neural networks and multiclass classification problems. Our method is implemented as an additional regularizer in Keras, a modular neural networks library written in Python, and evaluated in the benchmark data sets Reuters Newswire Topics Classification, IMDB database for binary sentiment classification, MNIST database of handwritten digits and CIFAR-10 data set for image classification.

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。

↑