Noise Sensitivity and Stability of Deep Neural Networks for Binary Classification

2023/08/18 by Johan Jonasson, Jonasson, Johan, Jeffrey E. Steif +3 · 1 citation
Computer Science · #Adversarial Robustness in Machine Learning #FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Algorithms #Neural Networks and Applications #Probability (math.PR)

paper · pdf · doi:10.48550/arxiv.2308.09374

openalex publication_date 2023/08/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

A first step is taken towards understanding often observed non-robustness phenomena of deep neural net (DNN) classifiers. This is done from the perspective of Boolean functions by asking if certain sequences of Boolean functions represented by common DNN models are noise sensitive or noise stable, concepts defined in the Boolean function literature. Due to the natural randomness in DNN models, these concepts are extended to annealed and quenched versions. Here we sort out the relation between these definitions and investigate the properties of two standard DNN architectures, the fully connected and convolutional models, when initiated with Gaussian weights.

Cited by

• Random Spiking Neural Networks are Stable and Spectrally Simple

Related

• CodNN -- Robust Neural Networks From Coded Classification
• The Resistance to Label Noise in K-NN and DNN Depends on its\n Concentration
• Robust Loss Functions under Label Noise for Deep Neural Networks
• On the Stability of Deep Networks
• Deep neural networks have an inbuilt Occam's razor