Cognitive Bias-Inspired Deep Robust Neural Networks Against Transfer-Based Attacks Considering Confidence Score

Cognitive Bias-Inspired Deep Robust Neural Networks Against Transfer-Based Attacks Considering Confidence Score

Transfer-based attacks, a type of adversarial attack, have become a growing threat in recent years with the proliferation of cloud services. Deep neural networks that exploit human cognitive bias (Loosely Symmetric-Deep Neural Network, LS-DNN) are known as a defensive technique against transfer-base...

Full description

Bibliographic Details
Main Authors: Yuuki Ogasawara, Hiroshi Sato, Masao Kubo
Format: Article
Language:English
Published: World Scientific Publishing 2024-02-01
Series:Vietnam Journal of Computer Science
Subjects:
Adversarial examples
transfer-based attacks
cognitive bias
neural networks
robustness
confidence score
Online Access:https://www.worldscientific.com/doi/10.1142/S2196888823500112

Internet

https://www.worldscientific.com/doi/10.1142/S2196888823500112

Similar Items