Homotopy Relaxation Training Algorithms for Infinite-Width Two-Layer ReLU Neural Networks

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yang, Yahong, Chen, Qipin, Hao, Wenrui
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866929528579817472
author Yang, Yahong
Chen, Qipin
Hao, Wenrui
author_facet Yang, Yahong
Chen, Qipin
Hao, Wenrui
contents In this paper, we present a novel training approach called the Homotopy Relaxation Training Algorithm (HRTA), aimed at accelerating the training process in contrast to traditional methods. Our algorithm incorporates two key mechanisms: one involves building a homotopy activation function that seamlessly connects the linear activation function with the ReLU activation function; the other technique entails relaxing the homotopy parameter to enhance the training refinement process. We have conducted an in-depth analysis of this novel method within the context of the neural tangent kernel (NTK), revealing significantly improved convergence rates. Our experimental results, especially when considering networks with larger widths, validate the theoretical conclusions. This proposed HRTA exhibits the potential for other activation functions and deep neural networks.
format Preprint
id arxiv_https___arxiv_org_abs_2309_15244
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Homotopy Relaxation Training Algorithms for Infinite-Width Two-Layer ReLU Neural Networks
Yang, Yahong
Chen, Qipin
Hao, Wenrui
Machine Learning
Optimization and Control
In this paper, we present a novel training approach called the Homotopy Relaxation Training Algorithm (HRTA), aimed at accelerating the training process in contrast to traditional methods. Our algorithm incorporates two key mechanisms: one involves building a homotopy activation function that seamlessly connects the linear activation function with the ReLU activation function; the other technique entails relaxing the homotopy parameter to enhance the training refinement process. We have conducted an in-depth analysis of this novel method within the context of the neural tangent kernel (NTK), revealing significantly improved convergence rates. Our experimental results, especially when considering networks with larger widths, validate the theoretical conclusions. This proposed HRTA exhibits the potential for other activation functions and deep neural networks.
title Homotopy Relaxation Training Algorithms for Infinite-Width Two-Layer ReLU Neural Networks
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2309.15244