On Need for Topology-Aware Generative Models for Manifold-Based Defenses

by   Uyeong Jang, et al.

ML algorithms or models, especially deep neural networks (DNNs), have shown significant promise in several areas. However, recently researchers have demonstrated that ML algorithms, especially DNNs, are vulnerable to adversarial examples (slightly perturbed samples that cause mis-classification). Existence of adversarial examples has hindered deployment of ML algorithms in safety-critical sectors, such as security. Several defenses for adversarial examples exist in the literature. One of the important classes of defenses are manifold-based defenses, where a sample is "pulled back" into the data manifold before classifying. These defenses rely on the manifold assumption (data lie in a manifold of lower dimension than the input space). These defenses use a generative model to approximate the input distribution. This paper asks the following question: do the generative models used in manifold-based defenses need to be topology-aware? Our paper suggests the answer is yes. We provide theoretical and empirical evidence to support our claim.


On Need for Topology Awareness of Generative Models

Manifold assumption in learning states that: the data lie approximately ...

A survey on Adversarial Attacks and Defenses in Text

Deep neural networks (DNNs) have shown an inherent vulnerability to adve...

Generating Semantic Adversarial Examples with Differentiable Rendering

Machine learning (ML) algorithms, especially deep neural networks, have ...

Analyzing Accuracy Loss in Randomized Smoothing Defenses

Recent advances in machine learning (ML) algorithms, especially deep neu...

Generative Models for Security: Attacks, Defenses, and Opportunities

Generative models learn the distribution of data from a sample dataset a...

Manifold Preserving Adversarial Learning

How to generate semantically meaningful and structurally sound adversari...

For Manifold Learning, Deep Neural Networks can be Locality Sensitive Hash Functions

It is well established that training deep neural networks gives useful r...