Deep Learning

Research Frontiers in Backdoor Attacks and Model Poisoning Defense

Investigation of attack vectors where malicious data corrupts model training and development of robust defenses against poisoning and backdoor vulnerabilities.

All Deep Learning PhD categories