Saved in:
Bibliographic Details
Main Authors: Fischer, Manfred M., Pitts, Joshua
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.13298
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910199968694272
author Fischer, Manfred M.
Pitts, Joshua
author_facet Fischer, Manfred M.
Pitts, Joshua
contents This paper investigates the relationship between convolutional neural network (CNN) topology and image recognition performance through a comparative study of the VGG, ResNet, and GoogLeNet architectural families. Utilizing a unified experimental framework, the study isolates the impact of depth from confounding implementation variables. A formal distinction is introduced between nominal depth ($D_{\mathrm{nom}}$), representing the physical layer count, and effective depth ($D_{\mathrm{eff}}$), an operational metric quantifying the expected number of sequential transformations. Empirical results demonstrate that architectures utilizing identity shortcuts or branching modules maintain optimization stability by decoupling $D_{\mathrm{eff}}$ from $D_{\mathrm{nom}}$. These findings suggest that effective depth serves as a superior framework for predicting scaling potential and practical trainability, ultimately indicating that architectural topology - rather than sheer layer volume - is the primary determinant of gradient health in deep learning models.
format Preprint
id arxiv_https___arxiv_org_abs_2602_13298
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Effective Depth Paradox: Evaluating the Relationship between Architectural Topology and Trainability in Deep CNNs
Fischer, Manfred M.
Pitts, Joshua
Computer Vision and Pattern Recognition
Artificial Intelligence
This paper investigates the relationship between convolutional neural network (CNN) topology and image recognition performance through a comparative study of the VGG, ResNet, and GoogLeNet architectural families. Utilizing a unified experimental framework, the study isolates the impact of depth from confounding implementation variables. A formal distinction is introduced between nominal depth ($D_{\mathrm{nom}}$), representing the physical layer count, and effective depth ($D_{\mathrm{eff}}$), an operational metric quantifying the expected number of sequential transformations. Empirical results demonstrate that architectures utilizing identity shortcuts or branching modules maintain optimization stability by decoupling $D_{\mathrm{eff}}$ from $D_{\mathrm{nom}}$. These findings suggest that effective depth serves as a superior framework for predicting scaling potential and practical trainability, ultimately indicating that architectural topology - rather than sheer layer volume - is the primary determinant of gradient health in deep learning models.
title The Effective Depth Paradox: Evaluating the Relationship between Architectural Topology and Trainability in Deep CNNs
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2602.13298