Jiaqi Li

Machine Learning Theory

Concentration Inequalities for Non-Convex Stochastic Gradient Descent (working paper)
Gabriel Clara, J. Li, Peter L. Bartlett
Finite-Sample Distribution Theory and Efficient Large-Scale Inference for Online Quantile Regression (working paper)
Ziyang Wei^*, J. Li^†, Lan Wang, Wei Biao Wu
(2026+) Online Inference for Federated Learning with Stochastic Controlled Averaging
Noah Geller^*, J. Li^†
Under review.
(2026) Asymptotics of Stochastic Gradient Descent with Dropout Regularization in Linear Models
J. Li^†, Johannes Schmidt-Hieber, Wei Biao Wu
The Journal of Machine Learning Research (JMLR), accepted.
[R code: Github 'Dropout_SGD']
(2025) Statistical Guarantees for High-Dimensional Stochastic Gradient Descent
J. Li^†, Zhipeng Lou, Johannes Schmidt-Hieber, Wei Biao Wu
The 39th Annual Conference on Neural Information Processing Systems (NeurIPS), accepted.
(2025) Gaussian Approximation and Concentration of Constant Learning-Rate Stochastic Gradient Descent
Ziyang Wei^*, J. Li^†, Zhipeng Lou, Wei Biao Wu
The 39th Annual Conference on Neural Information Processing Systems (NeurIPS), accepted.
(2025) Asymptotic Theory of SGD with a General Learning-Rate
Or Goldreich^*, Ziyang Wei^*, Soham Bonnerjee^*, J. Li, Wei Biao Wu
The 39th Annual Conference on Neural Information Processing Systems (NeurIPS), accepted.
(2025+) Central Limit Theorems and Online Inference for Quantiles by Stochastic Gradient Descent
Ziyang Wei^*, J. Li^†, Likai Chen, Wei Biao Wu
IEEE Transactions on Information Theory, revision submitted.
(2024+) Stochastic Gradient Descent: A Nonlinear Time Series Perspective
J. Li^†, Zhipeng Lou, Stefan Richter, Wei Biao Wu
Journal of the Royal Statistical Society Series B, under revision.
(2024+) L^p-Convergence of Stochastic Gradient Descent with a Mixture Learning Rate
Qianqian Lei^*, J. Li^†, Or Goldreich^*, Sung-Yi Lin^*, Wei Biao Wu
The Journal of Time Series Analysis, under review.
(2024) Probabilistic Guarantees of Stochastic Recursive Gradient in Non-Convex Finite Sum Problems
Yanjie Zhong, J. Li^†, Soumendra Lahiri
The 28th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD).
[Python code: Github 'Prob-SARAH-PyTorch']
(2023) Statistical Analysis of Adaptive Variance-Reduced EM Algorithms Under the Polyak-Łojasiewicz Condition
Yanjie Zhong, J. Li, Todd Kuffner
(2018) Design of Wi-Fi Indoor Positioning Model and Algorithm Based on Radial Basis Function with Optimal Shape Parameter
J. Li^†
IEEE 10th International Conference on Communication Software and Networks (ICCSN)
Oral presentation honored as "Best Talk", acceptance < 1%

Time Series

Inference for Quantile Change Points in High-Dimensional Time Series (working paper)
Wentao Hu^*, Qianqian Lei^*, J. Li^†, Likai Chen, Mengyu Xu
Adaptive Two-Way MOSUM: Testing for Multiscale Change Points in Spatial Time Series (working paper)
Hangcen Zou^*, J. Li^†, Michel F. C. Haddad, Likai Chen
(2024) L² Inference for Change Points in High-Dimensional Time Series via a Two-Way MOSUM
J. Li^†, Likai Chen, Weining Wang, Wei Biao Wu
The Annals of Statistics (AOS). 52(2): 602-627.
[R code: R package 'L2hdchange']
(2024) Simultaneous Inference of a Partially Linear Model in Time Series
J. Li^†, Likai Chen, Kun Ho Kim, Tianwei Zhou
The Journal of Time Series Analysis (JTSA). 46: 623-646.
(2022) Change-Point Detection for High-Dimensional Time Series with Cross-Sectional Dependence
J. Li, Likai Chen, Jia He, Maggie Cheng, Wei Biao Wu

Neuroimaging

(2026+) On the Predictive Value of Message Passing for Noisy Connectivity Matrices
Tommaso Castellani^*, J. Li, Muriah D. Wheelock, Rezwana Rosen Razzaque, Helmut Laufs, Hong Chen, Claire Donnat
Under review.
(2026+) Decoding In-Scanner Non-Rapid Eye Movement (NREM) Sleep in Adolescents from Resting-State Functional Connectivity Data
Rezwana Rosen Razzaque^*, Han Li, Jiaxin Cindy Tu, Hong Chen, Donna Dierker, Andrew Eck, Jim Pollaro, J. Li, Helmut Laufs, Benjamin P. Kay, Deanna M. Barch, Muriah D. Wheelock
Society for Developmental Cognitive Neuroscience (FLUX).
(2025+) Age-Mismatched Functional Network Models Limit Outcome Prediction Performance in Developing Populations
Wei Li^*, J. Li, Ari Segel, Chenyan Lu, Andy Eck, Robert Lunde, Jed T. Elison, JC Tu, Ashley N. Nielsen, Muriah D. Wheelock
Submitting to Developmental Cognitive Neuroscience.
(2025) Decoding Unintentional Sleep Patterns from Functional MRI Data
Rezwana Rosen Razzaque^*, Jim Pollaro, Hong Chen, Donna Dierker, J. Li, and Muriah D. Wheelock
Imaging Science Retreat.
(2024) Network Level Enrichment Provides a Framework for Biological Interpretation of Machine Learning Results
J. Li, Ari Segel, Xinyang Feng, Jiaxin Cindy Tu, Andy Eck, Kelsey King, Babatunde Adeyemo, Nicole R. Karcher, Likai Chen, Adam T. Eggebrecht, Muriah D. Wheelock
Network Neuroscience. 8(3): 762–790.
[MATLAB code: Github 'MachineLearning_NetworkLevelAnalysis']
(2024) The Generalizability of Cortical Area Parcellations across Early Childhood
JC Tu, M Myers, W Li, J. Li, X Wang, D Dierker, T Day, AZ Snyder, A Latham, JK Kenley, CM. Sobolewski, Y Wang, AK Labonte, E Feczko, O Kardan, LA Moore, CM Sylvester, DA Fair, JT Elison, BB Warner, DM Barch, CE Rogers, JL Luby, CD Smyser, EM Gordon, TO Laumann, AT Eggebrecht, MD Wheelock
Cerebral Cortex, 35(5):116.
[MATLAB code: Github 'AreaParcellationInfants']
(2023) Effects of functional network model definition on biomarker outcome prediction
Xinyang Feng^*, Wei Li^*, J. Li, Jiaxin C. Tu, Ari Segel, Xinyang Feng, Muriah D. Wheelock

Research

Research

Research

Machine Learning Theory

Time Series

Neuroimaging