arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.01033cs.LGcs.NE
收藏

通过权重置换训练的神经网络是通用逼近器

Neural Networks Trained by Weight Permutation are Universal Approximators

The Hong Kong Polytechnic University(香港理工大学);Research Institute for Smart Energy, The Hong Kong Polytechnic University(香港理工大学智慧能源研究院)

Yongqiang Cai, Gaohang Chen, Zhonghua Qiao

AI总结:

本文针对无参数约束的传统神经网络训练,提出基于权重置换的训练方法,证明其可引导ReLU网络逼近一维连续函数并验证回归有效性,为该方法提供理论支撑。

摘要:

通用逼近特性是神经网络成功的基础,传统上通过对参数无任何约束的网络训练来实现。然而,近期实验研究提出了一种基于置换的新型训练方法,该方法在不修改权重精确值的情况下展现出了理想的分类性能。本文通过证明该置换训练方法能够引导ReLU网络逼近一维连续函数,为其提供了理论保证。我们的数值结果进一步验证了该方法在多种初始化设置下的回归任务中的有效性。权重置换过程中的显著发现表明,置换训练可为描述网络学习行为提供一种创新工具。

英文摘要:

The universal approximation property is fundamental to the success of neural networks, and has traditionally been achieved by training networks without any constraints on their parameters. However, recent experimental research proposed a novel permutation-based training method, which exhibited a desired classification performance without modifying the exact weight values. In this paper, we provide a theoretical guarantee of this permutation training method by proving its ability to guide a ReLU network to approximate one-dimensional continuous functions. Our numerical results further validate this method's efficiency in regression tasks with various initializations. The notable observations during weight permutation suggest that permutation training can provide an innovative tool for describing network learning behavior.

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。

↑