Saved in:
Bibliographic Details
Main Authors: Ge, Shufei, Wang, Shijia, Elliott, Lloyd
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.05586
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909985619836928
author Ge, Shufei
Wang, Shijia
Elliott, Lloyd
author_facet Ge, Shufei
Wang, Shijia
Elliott, Lloyd
contents Neural networks have shown state-of-the-art performances in various classification and regression tasks. Rectified linear units (ReLU) are often used as activation functions for the hidden layers in a neural network model. In this article, we establish the connection between the Poisson hyperplane processes (PHP) and two-layer ReLU neural networks. We show that the PHP with a Gaussian prior is an alternative probabilistic representation to a two-layer ReLU neural network. In addition, we show that a two-layer neural network constructed by PHP is scalable to large-scale problems via the decomposition propositions. Finally, we propose an annealed sequential Monte Carlo algorithm for Bayesian inference. Our numerical experiments demonstrate that our proposed method outperforms the classic two-layer ReLU neural network. The implementation of our proposed model is available at https://github.com/ShufeiGe/Pois_Relu.git.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05586
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Poisson Hyperplane Processes with Rectified Linear Units
Ge, Shufei
Wang, Shijia
Elliott, Lloyd
Machine Learning
Methodology
Neural networks have shown state-of-the-art performances in various classification and regression tasks. Rectified linear units (ReLU) are often used as activation functions for the hidden layers in a neural network model. In this article, we establish the connection between the Poisson hyperplane processes (PHP) and two-layer ReLU neural networks. We show that the PHP with a Gaussian prior is an alternative probabilistic representation to a two-layer ReLU neural network. In addition, we show that a two-layer neural network constructed by PHP is scalable to large-scale problems via the decomposition propositions. Finally, we propose an annealed sequential Monte Carlo algorithm for Bayesian inference. Our numerical experiments demonstrate that our proposed method outperforms the classic two-layer ReLU neural network. The implementation of our proposed model is available at https://github.com/ShufeiGe/Pois_Relu.git.
title Poisson Hyperplane Processes with Rectified Linear Units
topic Machine Learning
Methodology
url https://arxiv.org/abs/2601.05586