Interactive Analysis of CNN Robustness

10/14/2021
by   Stefan Sietzen, et al.
21

While convolutional neural networks (CNNs) have found wide adoption as state-of-the-art models for image-related tasks, their predictions are often highly sensitive to small input perturbations, which the human vision is robust against. This paper presents Perturber, a web-based application that allows users to instantaneously explore how CNN activations and predictions evolve when a 3D input scene is interactively perturbed. Perturber offers a large variety of scene modifications, such as camera controls, lighting and shading effects, background modifications, object morphing, as well as adversarial attacks, to facilitate the discovery of potential vulnerabilities. Fine-tuned model versions can be directly compared for qualitative evaluation of their robustness. Case studies with machine learning experts have shown that Perturber helps users to quickly generate hypotheses about model vulnerabilities and to qualitatively compare model behavior. Using quantitative analyses, we could replicate users' insights with other CNN architectures and input images, yielding new insights about the vulnerability of adversarially trained models.

READ FULL TEXT

page 2

page 5

page 6

page 7

page 8

page 9

page 10

page 11

research
03/06/2023

Visual Analytics of Neuron Vulnerability to Adversarial Attacks on Convolutional Neural Networks

Adversarial attacks on a convolutional neural network (CNN) – injecting ...
research
03/14/2021

BreakingBED – Breaking Binary and Efficient Deep Neural Networks by Adversarial Attacks

Deploying convolutional neural networks (CNNs) for embedded applications...
research
03/01/2021

Brain Programming is Immune to Adversarial Attacks: Towards Accurate and Robust Image Classification using Symbolic Learning

In recent years, the security concerns about the vulnerability of Deep C...
research
07/17/2020

Neural Networks with Recurrent Generative Feedback

Neural networks are vulnerable to input perturbations such as additive n...
research
03/31/2022

Towards Robust Rain Removal Against Adversarial Attacks: A Comprehensive Benchmark Analysis and Beyond

Rain removal aims to remove rain streaks from images/videos and reduce t...
research
06/03/2015

Understanding deep features with computer-generated imagery

We introduce an approach for analyzing the variation of features generat...
research
01/20/2023

Image Memorability Prediction with Vision Transformers

Behavioral studies have shown that the memorability of images is similar...

Please sign up or login with your details

Forgot password? Click here to reset