The geometry of neural network decision boundaries

Artificial neural networks (ANNs) are built using well-understood mathematical operations, yet their complex nature poses significant challenges to intuitively understanding their decision-making processes. This work introduces a method for visualizing and measuring ANN decision boundary geometry directly in the high-dimensional input space, avoiding distortions found with traditional projection-based techniques. We analyze boundary curvatures and distances in diverse inputs, architectures, and training methods. Boundary distances grow smoothly and slowly with consistent curvature as the size of the subspace increases, suggesting that decision boundaries are close to input samples within large, continuous subspaces. We also quantify how much boundary distances and curvatures reduce near dataset samples after adversarial training. Training methodology has a significantly stronger impact on boundary distance and curvature than architecture or test-set accuracy. These findings are contextualized within prior work, supporting or refuting several previously proposed theories of boundary behavior. Our analytically and empirically supported methodology lays the foundation for future studies to improve network interpretability and understanding.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-18
DOI
https://doi.org/10.48550/arxiv.2611.24557
Primary Topic
Explainable Artificial Intelligence (XAI)
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

The geometry of neural network decision boundaries

Dylan M. Paiton, Dirk Schultheiss, Zac Cranko, Matthias Bethge et al.
arXiv (Cornell University)
Explainable Artificial Intelligence (XAI)
preprint

The geometry of neural network decision boundaries

Dylan M. Paiton, Dirk Schultheiss, Zac Cranko, Matthias Bethge, Matthias Kümmerer
preprint en

Abstract

Artificial neural networks (ANNs) are built using well-understood mathematical operations, yet their complex nature poses significant challenges to intuitively understanding their decision-making processes. This work introduces a method for visualizing and measuring ANN decision boundary geometry directly in the high-dimensional input space, avoiding distortions found with traditional projection-based techniques. We analyze boundary curvatures and distances in diverse inputs, architectures, and training methods. Boundary distances grow smoothly and slowly with consistent curvature as the size of the subspace increases, suggesting that decision boundaries are close to input samples within large, continuous subspaces. We also quantify how much boundary distances and curvatures reduce near dataset samples after adversarial training. Training methodology has a significantly stronger impact on boundary distance and curvature than architecture or test-set accuracy. These findings are contextualized within prior work, supporting or refuting several previously proposed theories of boundary behavior. Our analytically and empirically supported methodology lays the foundation for future studies to improve network interpretability and understanding.

arXiv (Cornell University)
Explainable Artificial Intelligence (XAI)
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon