Towards Explaining Adversarial Examples Phenomenon in Artificial Neural Networks
arXiv:2107.10599 · doi:10.1109/ICPR48806.2021.9412367
Abstract
In this paper, we study the adversarial examples existence and adversarial training from the standpoint of convergence and provide evidence that pointwise convergence in ANNs can explain these observations. The main contribution of our proposal is that it relates the objective of the evasion attacks and adversarial training with concepts already defined in learning theory. Also, we extend and unify some of the other proposals in the literature and provide alternative explanations on the observations made in those proposals. Through different experiments, we demonstrate that the framework is valuable in the study of the phenomenon and is applicable to real-world problems.
submitted to 25th International Conference on Pattern Recognition (ICPR)
References in corpus (5)
- Explaining and Harnessing Adversarial Examples
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- A Boundary Tilting Persepective on the Phenomenon of Adversarial Examples
- A Simple Explanation for the Existence of Adversarial Examples with Small Hamming Distance
- Simple Physical Adversarial Examples against End-to-End Autonomous Driving Models