A5041729890- Adversarial Robustness in Machine Learning
- Anomaly Detection Techniques and Applications
- Digital Media Forensic Detection
- Generative Adversarial Networks and Image Synthesis
- Advanced Neural Network Applications
- Advanced Malware Detection Techniques
- Image Processing Techniques and Applications
- Bacillus and Francisella bacterial research
Anhui University
2023-2025
Deep Neural Networks (DNNs) have recently made significant progress in many fields. However, studies shown that DNNs are vulnerable to adversarial examples, where imperceptible perturbations can greatly mislead even if the full underlying model parameters not accessible. Various defense methods been proposed, such as feature compression and gradient masking. numerous proven previous create detection or against certain attacks, which renders method ineffective face of latest unknown attack...
Adding perturbations to images can mislead classification models produce incorrect results. Recently, researchers exploited adversarial protect image privacy from retrieval by intelligent models. However, adding destroys the original data, making useless in digital forensics and other fields. To prevent illegal or unauthorized access sensitive data such as human faces without impeding legitimate users, use of reversible attack techniques is increasing. The be recovered its examples. existing...