Assessing and Improving DNN Robustness against Adversarial Examples from the Perspective of Fully Connected Layers
- 25 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. PP (15566013) , 1
Abstract
Recent studies show that deep neural networks are extremely vulnerable, especially for adversarial examples of image classification models. However, existing defenses suffer from limited adaptability across attacks, an unfavorable tradeoff between clean accuracy and robustness, and substantial training-time overhead. To tackle these problems, we present a novel component, named the redundant fully connected layer, which can be combined with existing model backbones in a pluggable manner. Specifically, we design a tailor-made loss function for it that leverages cosine similarity to maximize the difference and diversity of multiple fully connected parts. We conduct extensive experiments against 12 representative attacks (white-box and black-box), based on two popular datasets. The empirical evaluations show that our scheme realizes significant outcomes against various attacks with negligible additional training overhead, while hardly bringing collateral damage for clean-instance accuracy.Keywords
This publication has 41 references indexed in Scilit:
- Square Attack: A Query-Efficient Black-Box Adversarial Attack via Random Search2020
- Feature Distillation: DNN-Oriented JPEG Compression Against Adversarial Examples2019
- Improving Transferability of Adversarial Examples With Input Diversity2019
- Towards Robust Neural Networks via Random Self-ensemble2018
- Boosting Adversarial Attacks with Momentum2018
- Densely Connected Convolutional Networks2017
- DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks2016
- Deep Residual Learning for Image Recognition2016
- Rethinking the Inception Architecture for Computer Vision2016
- ImageNet: A large-scale hierarchical image database2009