Yuanzhi Li is a researcher in the field of machine learning and artificial intelligence. Li's primary research area is deep learning theory, with a focus on understanding the hierarchical feature learning process in neural networks and how it differs from shallow learning methods. Li has authored or co-authored over 50 research papers with over 5600 citations.
Li received a Ph.D. in computer science from Princeton University, USA, and a Bachelor of Computer Science and Mathematics from Tsinghua University, China.
Li is currently an Affiliated Assistant Professor of Machine Learning at MBZUAI. Prior to joining MBZUAI, Li was a postdoctoral researcher at Stanford University and an assistant professor in the Carnegie Mellon University (CMU) Department of Machine Learning. Li has also held positions at the University of Toronto, Canada, and the Institute for Advanced Study at Princeton, USA.
In 2023, Li was selected for a prestigious Sloan Research Fellowship in computer science by the Alfred P. Sloan Foundation. The Fellowship is awarded to early-career researchers who demonstrate creativity, innovation, and outstanding research accomplishments. Li was one of only 22 scholars selected in computer science for 2023.
Li's research interests include deep learning, reinforcement learning, pre-training/foundation models, robustness, non-convex optimization, distributed optimization, and high-dimensional statistics. Li has published numerous papers in these areas, including:
Yuanzhi Li is a prominent researcher in the field of machine learning and artificial intelligence, with a focus on deep learning theory. Li has made significant contributions to the understanding of neural networks and their optimization algorithms, and has received recognition for their research accomplishments.
Youtube Title: Yuanzhi Li | Physics of Language Models: Knowledge Storage, Extraction, and Manipulation
Youtube Link: link
Youtube Channel Name: Harvard CMSA
Youtube Channel Link: https://www.youtube.com/@harvardcmsa7486
Yuanzhi Li | Physics of Language Models: Knowledge Storage, Extraction, and Manipulation
Youtube Title: Understanding Over-parametrization Through Matrix Sensing
Youtube Link: link
Youtube Channel Name: Microsoft Research
Youtube Channel Link: https://www.youtube.com/@MicrosoftResearch
Understanding Over-parametrization Through Matrix Sensing
Youtube Title: AI Talks | Understanding the mixture of the expert layer in Deep Learning | MBZUAI
Youtube Link: link
Youtube Channel Name: MBZUAI
Youtube Channel Link: https://www.youtube.com/@MBZUAI
AI Talks | Understanding the mixture of the expert layer in Deep Learning | MBZUAI
Youtube Title: Deep Learning: From an Alchemist to a Theoretical Alchemist
Youtube Link: link
Youtube Channel Name: Machine Learning Department at CMU
Youtube Channel Link: https://www.youtube.com/@mldcmu
Deep Learning: From an Alchemist to a Theoretical Alchemist
Youtube Title: Operator Scaling via Geodesically Convex Optimization, Invariant… (Continued) - Yuanzhi Li
Youtube Link: link
Youtube Channel Name: Institute for Advanced Study
Youtube Channel Link: https://www.youtube.com/@videosfromIAS
Operator Scaling via Geodesically Convex Optimization, Invariant... (Continued) - Yuanzhi Li
Youtube Title: Operator Scaling via Geodesically Convex Optimization, Invariant Theory… - Yuanzhi Li
Youtube Link: link
Youtube Channel Name: Institute for Advanced Study
Youtube Channel Link: https://www.youtube.com/@videosfromIAS
Operator Scaling via Geodesically Convex Optimization, Invariant Theory... - Yuanzhi Li
Youtube Title: CS201 YUANZHI LI MARCH 2 2021
Youtube Link: link
Youtube Channel Name: UCLA Computer Science
Youtube Channel Link: https://www.youtube.com/@uclacomputerscience1
CS201 YUANZHI LI MARCH 2 2021
Youtube Title: Operator Scaling via Geodesically Convex Optimization, Invariant Theory… - Yuanzhi Li
Youtube Link: link
Youtube Channel Name: Institute for Advanced Study
Youtube Channel Link: https://www.youtube.com/@videosfromIAS
Operator Scaling via Geodesically Convex Optimization, Invariant Theory... - Yuanzhi Li
Youtube Title: Feature purification: How adversarial training can perform robust deep learning - Yuanzhi Li
Youtube Link: link
Youtube Channel Name: Institute for Advanced Study
Youtube Channel Link: https://www.youtube.com/@videosfromIAS
Feature purification: How adversarial training can perform robust deep learning - Yuanzhi Li
Youtube Title: The Tiny Model Revolution with Ronen Eldan and Yuanzhi Li of Microsoft Research
Youtube Link: link
Youtube Channel Name: Cognitive Revolution "How AI Changes Everything"
Youtube Channel Link: https://www.youtube.com/@CognitiveRevolutionPodcast
The Tiny Model Revolution with Ronen Eldan and Yuanzhi Li of Microsoft Research
Youtube Title: Li Yuanzhi video for EBAC
Youtube Link: link
Youtube Channel Name: spancer lee
Youtube Channel Link: https://www.youtube.com/@spancerlee3153
Li Yuanzhi video for EBAC
Youtube Title: Jason Yuanzhi Li and Bonnie Yuanhui Li
Youtube Link: link
Youtube Channel Name: Bonnie_li
Youtube Channel Link: https://www.youtube.com/@xinnannancy
Jason Yuanzhi Li and Bonnie Yuanhui Li
Youtube Title: Backward Feature Correction: How Deep Learning Performs Deep Learning (May 2020 by Yuanzhi Li)
Youtube Link: link
Youtube Channel Name: Zeyuan Allen-Zhu
Youtube Channel Link: https://www.youtube.com/@zhuzeyuan
Backward Feature Correction: How Deep Learning Performs Deep Learning (May 2020 by Yuanzhi Li)
Youtube Title: An Instant Archive?_Su Diyue, Zheng Yuanzhi, Li Zihan
Youtube Link: link
Youtube Channel Name: Skunkworks
Youtube Channel Link: https://www.youtube.com/@skunkworks6868
An Instant Archive?_Su Diyue, Zheng Yuanzhi, Li Zihan
Youtube Title: Spotlight Talks - Various
Youtube Link: link
Youtube Channel Name: Institute for Advanced Study
Youtube Channel Link: https://www.youtube.com/@videosfromIAS
Spotlight Talks - Various
Youtube Title: Michael Cohen and Self-concordant Barriers over L_p Balls
Youtube Link: link
Youtube Channel Name: Simons Institute
Youtube Channel Link: https://www.youtube.com/@SimonsInstituteTOC
Michael Cohen and Self-concordant Barriers over L_p Balls