Shuai Zhang
Assistant Professor, Data Science
2117 Guttenberg Information Technologies Center (GITC)
About Me
I am an assistant professor in the Department of Data Science, New Jersey Institute of Technology (NJIT), starting at Fall 2023. I earned my Ph.D. in Electrical and Computer Engineering from Rensselaer Polytechnic Institute under the supervision of Dr. Meng Wang. Previously, I was fortunately to work with IBM Thomas J. Watson Research Center and MIT-IBM Watson AI Lab.
Education
Ph.D.; Rensselaer Polytechnic Institute; Electrical and Computer Engineering; 2021
B.E.; University of Science and Technology of China; 2016
B.E.; University of Science and Technology of China; 2016
Past Courses
CS 732: ADVANCED MACHINE LEARNING
DS 400: SCIENTIFIC FOUNDATION OF MACHINE LEARNING
DS 675: MACHINE LEARNING
DS 400: SCIENTIFIC FOUNDATION OF MACHINE LEARNING
DS 675: MACHINE LEARNING
Research Interests
My research focuses on the learning theory of emerging artificial intelligence technologies and the structural properties of modern models, with an emphasis on understanding their implications for training dynamics, optimization, and generalization. My research aims to develop principled theoretical frameworks that explain how architectural structure, optimization geometry, and data properties interact to shape the performance, efficiency, and reliability of AI systems.
Conference Paper
"A Theoretical Analysis of Mamba’s Training Dynamics: Filtering Relevant Features for Generalization in State Space Models"
ICLR 2026.
"Theoretical Analysis of Contrastive Learning under Imbalanced Data: From Training Dynamics to a Pruning Solution"
ICLR 2026.
"Theoretical Guarantees and Training Dynamics of Contrastive Learning: How Misaligned Data Influence Feature Purity"
2025.
"When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers"
2025.
"Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning"
ICML 2024, 2024.
ICLR 2026.
"Theoretical Analysis of Contrastive Learning under Imbalanced Data: From Training Dynamics to a Pruning Solution"
ICLR 2026.
"Theoretical Guarantees and Training Dynamics of Contrastive Learning: How Misaligned Data Influence Feature Purity"
2025.
"When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers"
2025.
"Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning"
ICML 2024, 2024.
SHOW MORE
Journal Article
Shusen Jing, Anlan Yu, Shuai Zhang, Songyang Zhang.
2024. "FedSC: Provable Federated Self-supervised Learning with Spectral Contrastive Objective over Non-iid Data."
ICML, 2024 .
Shuai Zhang, Heshan Devaka Fernando, Miao Liu, Keerthiram Murugesan, Songtao Lu, Pin-Yu Chen, Tianyi Chen, Meng Wang. 2024. "SF-DQN: Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning." ICML, 2024 .
Shuai Zhang, Meng Wang. 2019. "Correction of corrupted columns through fast robust Hankel matrix completion." IEEE Transactions on Signal Processing , vol. 67 , no. 10 , pp. 2580--2594.
Shuai Zhang, Yingshuai Hao, Meng Wang, Joe H Chow. 2018. "Multichannel Hankel matrix completion through nonconvex optimization." IEEE Journal of Selected Topics in Signal Processing , vol. 12 , no. 4 , pp. 617--632.
Shuai Zhang, Heshan Devaka Fernando, Miao Liu, Keerthiram Murugesan, Songtao Lu, Pin-Yu Chen, Tianyi Chen, Meng Wang. 2024. "SF-DQN: Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning." ICML, 2024 .
Shuai Zhang, Meng Wang. 2019. "Correction of corrupted columns through fast robust Hankel matrix completion." IEEE Transactions on Signal Processing , vol. 67 , no. 10 , pp. 2580--2594.
Shuai Zhang, Yingshuai Hao, Meng Wang, Joe H Chow. 2018. "Multichannel Hankel matrix completion through nonconvex optimization." IEEE Journal of Selected Topics in Signal Processing , vol. 12 , no. 4 , pp. 617--632.