Score: 1

Topological Signatures of ReLU Neural Network Activation Patterns

Published: October 14, 2025 | arXiv ID: 2510.12700v1

By: Vicente Bosca , Tatum Rask , Sunia Tanweer and more

BigTech Affiliations: University of Washington

Potential Business Impact:

Finds patterns in how computer brains learn.

Business Areas:
Image Recognition Data and Analytics, Software

This paper explores the topological signatures of ReLU neural network activation patterns. We consider feedforward neural networks with ReLU activation functions and analyze the polytope decomposition of the feature space induced by the network. Mainly, we investigate how the Fiedler partition of the dual graph and show that it appears to correlate with the decision boundary -- in the case of binary classification. Additionally, we compute the homology of the cellular decomposition -- in a regression task -- to draw similar patterns in behavior between the training loss and polyhedral cell-count, as the model is trained.

Country of Origin
🇺🇸 United States

Page Count
14 pages

Category
Computer Science:
Machine Learning (CS)