T-AI

WELCOME TO T-AI

Trustworthy Artificial Intelligence (T-AI) lab is committed to advancing the frontiers of machine learning and security research, and applying breakthroughs to real-world problems.

Research for Fun:

• Trustworthy ML：With the goal of building robust, secure machine learning models, we explore topics such as adversarial attack/defense, backdoor attack/defense, robust and interpretable machine learning.
• Privacy Protection：We prioritize privacy protection for ML systems by researching techniques like differential privacy, secure multi-party computation, and homomorphic encryption. We also monitor new privacy issues like membership inference and model extraction, etc.

Research for World:

• AI for Driving: We make self-driving cars safer by creating smart technology that helps them "see" better, make better decisions, and follow laws. We also test their abilities and make sure they're safe and ethical to use.
• AI for Medicine: We collaborate with healthcare providers to develop precise and efficient AI algorithms for medical tasks like imaging and drug discovery, with a particular emphasis on nuclear medicine at present.