Recurrent Attention Model with Log-Polar Mapping is Robust against Adversarial Attacks

2020/02/13 by Taro Kiritani, Koji Ono, Kiritani, Taro +1
Computer Science · Engineering · #Advanced Memory and Neural Computing #Advanced Neural Network Applications #Adversarial Robustness in Machine Learning #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2002.05388

openalex publication_date 2020/02/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Convolutional neural networks are vulnerable to small ℓ^p adversarial attacks, while the human visual system is not. Inspired by neural networks in the eye and the brain, we developed a novel artificial neural network model that recurrently collects data with a log-polar field of view that is controlled by attention. We demonstrate the effectiveness of this design as a defense against SPSA and PGD adversarial attacks. It also has beneficial properties observed in the animal visual system, such as reflex-like pathways for low-latency inference, fixed amount of computation independent of image size, and rotation and scale invariance. The code for experiments is available at https://gitlab.com/exwzd-public/kiritani_ono₂020.

Citations

• Adam: A Method for Stochastic Optimization
• ImageNet Large Scale Visual Recognition Challenge
• The eye-movements of the mantis shrimp Odontodactylus scyllarus (Crustacea: Stomatopoda)
• Towards Deep Learning Models Resistant to Adversarial Attacks
• Spatially Transformed Adversarial Examples
• Robustness of Rotation-Equivariant Networks to Adversarial Perturbations
• advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorch

Related

• Towards Robust Image Classification Using Sequential Attention Models
• Human Eyes Inspired Recurrent Neural Networks are More Robust Against Adversarial Noises
• On Inherent Adversarial Robustness of Active Vision Systems
• Attention-Based Real-Time Defenses for Physical Adversarial Attacks in Vision Applications
• Advances in adversarial attacks and defenses in computer vision: A\n survey