Analyzing the Noise Robustness of Deep Neural Networks
This addresses the risk of applying DNNs to safety-critical applications like driverless cars, but it is incremental as it builds on existing methods for analyzing adversarial examples.
The paper tackles the vulnerability of deep neural networks to adversarial examples by presenting a visual analytics approach to explain the primary cause of wrong predictions, demonstrating its promise through two case studies.
Deep neural networks (DNNs) are vulnerable to maliciously generated adversarial examples. These examples are intentionally designed by making imperceptible perturbations and often mislead a DNN into making an incorrect prediction. This phenomenon means that there is significant risk in applying DNNs to safety-critical applications, such as driverless cars. To address this issue, we present a visual analytics approach to explain the primary cause of the wrong predictions introduced by adversarial examples. The key is to analyze the datapaths of the adversarial examples and compare them with those of the normal examples. A datapath is a group of critical neurons and their connections. To this end, we formulate the datapath extraction as a subset selection problem and approximately solve it based on back-propagation. A multi-level visualization consisting of a segmented DAG (layer level), an Euler diagram (feature map level), and a heat map (neuron level), has been designed to help experts investigate datapaths from the high-level layers to the detailed neuron activations. Two case studies are conducted that demonstrate the promise of our approach in support of explaining the working mechanism of adversarial examples.