Backdoor Attacks and Defenses in AI Systems - Seminar Series

Explore the critical topic of backdoor attacks and defenses in AI systems through this informative seminar presented by Dr. Tuan Anh Tran, a former Applied Scientist at Amazon Rekognition. Delve into the security threats posed by backdoor attacks in deep learning, particularly in the context of using pre-trained weights. Gain insights into how backdoor models can function normally with clean data but exhibit malicious behavior when triggered by specific patterns. Learn from Dr. Tran's extensive experience in facial image processing and computer vision as he shares his expertise on this crucial aspect of AI security. Benefit from his background in facial image analysis and his academic achievements, including honors such as Vietnam Talents 2010 and the Vietnam Education Foundation fellowship.