Defining Neural Network Architecture through Polytope Structures of Dataset

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Sangmin, Mammadov, Abbas, Ye, Jong Chul
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916265514237952
author Lee, Sangmin
Mammadov, Abbas
Ye, Jong Chul
author_facet Lee, Sangmin
Mammadov, Abbas
Ye, Jong Chul
contents Current theoretical and empirical research in neural networks suggests that complex datasets require large network architectures for thorough classification, yet the precise nature of this relationship remains unclear. This paper tackles this issue by defining upper and lower bounds for neural network widths, which are informed by the polytope structure of the dataset in question. We also delve into the application of these principles to simplicial complexes and specific manifold shapes, explaining how the requirement for network width varies in accordance with the geometric complexity of the dataset. Moreover, we develop an algorithm to investigate a converse situation where the polytope structure of a dataset can be inferred from its corresponding trained neural networks. Through our algorithm, it is established that popular datasets such as MNIST, Fashion-MNIST, and CIFAR10 can be efficiently encapsulated using no more than two polytopes with a small number of faces.
format Preprint
id arxiv_https___arxiv_org_abs_2402_02407
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Defining Neural Network Architecture through Polytope Structures of Dataset
Lee, Sangmin
Mammadov, Abbas
Ye, Jong Chul
Machine Learning
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
Current theoretical and empirical research in neural networks suggests that complex datasets require large network architectures for thorough classification, yet the precise nature of this relationship remains unclear. This paper tackles this issue by defining upper and lower bounds for neural network widths, which are informed by the polytope structure of the dataset in question. We also delve into the application of these principles to simplicial complexes and specific manifold shapes, explaining how the requirement for network width varies in accordance with the geometric complexity of the dataset. Moreover, we develop an algorithm to investigate a converse situation where the polytope structure of a dataset can be inferred from its corresponding trained neural networks. Through our algorithm, it is established that popular datasets such as MNIST, Fashion-MNIST, and CIFAR10 can be efficiently encapsulated using no more than two polytopes with a small number of faces.
title Defining Neural Network Architecture through Polytope Structures of Dataset
topic Machine Learning
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
url https://arxiv.org/abs/2402.02407