Some Geometrical and Topological Properties of DNNs' Decision Boundaries

2020/03/07 by Bo Liu, Liu, Bo, Mengya Shen +1
Computer Science · Mathematics · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #Domain Adaptation and Few-Shot Learning #cs.LG #stat.ML

paper · pdf · doi:10.48550/arxiv.2003.03687

arxiv created 2021/04/16 · arxiv updated 2021/04/19

Abstract

Geometry and topology of decision regions are closely related with classification performance and robustness against adversarial attacks. In this paper, we use differential geometry to theoretically explore the geometrical and topological properties of decision regions produced by deep neural networks (DNNs). The goal is to obtain some geometrical and topological properties of decision boundaries for given DNN models, and provide some principled guidance to design and regularization of DNNs. First, we present the curvatures of decision boundaries in terms of network parameters, and give sufficient conditions on network parameters for producing flat or developable decision boundaries. Based on the Gauss-Bonnet-Chern theorem in differential geometry, we then propose a method to compute the Euler characteristics of compact decision boundaries, and verify it with experiments.

Citations

• A Topological Regularizer for Classifiers via Persistent Homology

Related

• Classification regions of deep neural networks
• Investigating Decision Boundaries of Trained Neural Networks
• Topological Data Analysis of Decision Boundaries with Application to Model Selection
• A Differential Topological View of Challenges in Learning with Feedforward Neural Networks
• Characterizing the Decision Boundary of Deep Neural Networks