A Variational Inequality Model for Learning Neural Networks
A Variational Inequality Model for Learning Neural Networks
复制标题
DOI：
10.1109/icassp49357.2023.10095688
复制
发表时间：
2022-10
期刊：
影响因子：
--
通讯作者：
P. Combettes;J. Pesquet;A. Repetti
中科院分区：
文献类型：
--
作者：
P. Combettes;J. Pesquet;A. Repetti
神经网络已经成为解决信号和图像处理问题的普遍工具，它们的性能往往优于标准方法。然而，在许多应用中，训练神经网络的层是一项具有挑战性的任务。普遍的训练过程包括最小化高度非凸目标的基础上的数据集的巨大尺寸。在这方面，目前的方法不能保证产生全球性的解决办法。我们提出了一种替代方法，放弃了优化框架，并采用变分不等式形式主义。相关算法保证迭代收敛到变分不等式的真解，它具有一个有效的块迭代结构。数值应用。
Neural networks have become ubiquitous tools for solving signal and image processing problems, and they often outperform standard approaches. Nevertheless, training the layers of a neural network is a challenging task in many applications. The prevalent training procedure consists of minimizing highly non-convex objectives based on data sets of huge dimension. In this context, current methodologies are not guaranteed to produce global solutions. We present an alternative approach which foregoes the optimization framework and adopts a variational inequality formalism. The associated algorithm guarantees convergence of the iterates to a true solution of the variational inequality and it possesses an efficient block-iterative structure. A numerical application is presented.