Jiantao Jiao

Jiantao Jiao is an Adjunct Assistant Professor in the Department of Electrical Engineering (EE) at the University of California, Berkeley. His research spans a broad range of areas including Artificial Intelligence (AI), Information, Data, Network, and Communication Sciences (IDNCS), Control, Intelligent Systems, and Robotics (CIR), Theory (THY), Signal Processing (SP), Operating Systems & Networking (OSNT), Database Management Systems (DBMS), Cyber-Physical Systems and Design Automation (CPSDA), Security (SEC), Power and Energy (ENE), and Programming Systems (PS). He is affiliated with several research centers at UC Berkeley, including the Berkeley Artificial Intelligence Research Lab (BAIR), the Berkeley Center for Responsible, Decentralized Intelligence (RDI), the Berkeley Laboratory for Information and System Sciences (BLISS), the Center for the Theoretical Foundations of Learning, Inference, Information, Intelligence, Mathematics and Microeconomics at Berkeley (CLIMB), and the Simons Institute for the Theory of Computing (SITC).

Jiantao Jiao is a Director of Research and Distinguished Scientist at NVIDIA, and a Professor of EECS and Stat at UC Berkeley. His research focuses on Generative AI, Large Language Models, Reinforcement Learning, Machine Learning, and AI security. His work explores the theoretical foundations of AI systems, including robustness-accuracy trade-offs, offline reinforcement learning, imitation learning, and secure aggregation in federated learning. He has contributed to advancing principled reinforcement learning with human feedback, minimax estimation of functionals, and improving language model helpfulness and harmlessness through techniques like RLAIF. His research emphasizes the development of reliable, secure, and efficient AI systems.