Gradient Band-based Adversarial Training for Generalized Attack Immunity of A3C Path Finding

Abstract

As adversarial attacks pose a serious threat to the security of AI system in practice, such attacks have been extensively studied in the context of computer vision applications. However, few attentions have been paid to the adversarial research on automatic path finding. In this paper, we show dominant adversarial examples are effective when targeting A3C path finding, and design a Common Dominant Adversarial Examples Generation Method (CDG) to generate dominant adversarial examples against any given map. In addition, we propose Gradient Band-based Adversarial Training, which trained with a single randomly choose dominant adversarial example without taking any modification, to realize the "1:N" attack immunity for generalized dominant adversarial examples. Extensive experimental results show that, the lowest generation precision for CDG algorithm is 91.91%, and the lowest immune precision for Gradient Band-based Adversarial Training is 93.89%, which can prove that our method can realize the generalized attack immunity of A3C path finding with a high confidence.

25 pages 14 figures

References in corpus (15)

• Playing Atari with Deep Reinforcement Learning
• Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networks
• Delving into Transferable Adversarial Examples and Black-box Attacks
• Achieving Human Parity in Conversational Speech Recognition
• Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients
• Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning
• Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey
• Towards Proving the Adversarial Robustness of Deep Neural Networks
• Fast Feature Fool: A data independent approach to universal adversarial perturbations
• Extending Defensive Distillation
• DeepCloak: Masking Deep Neural Network Models for Robustness Against Adversarial Samples
• Cascade Adversarial Machine Learning Regularized with a Unified Embedding
• On the Robustness of Convolutional Neural Networks to Internal Architecture and Weight Perturbations
• Learning Adversary-Resistant Deep Neural Networks
• Defense against Universal Adversarial Perturbations

Cited by in corpus (2)

• Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observations
• A Training-based Identification Approach to VIN Adversarial Examples