Thumbnail

Jason Lee

Assistant Professor of Data Science and Operations
Building
BRI
Room / Office
307F

Jason Lee is an assistant professor in Data Sciences and Operations at the University of Southern California. Prior to that, he was a postdoctoral researcher at UC Berkeley working with Michael Jordan. Jason received his PhD at Stanford University advised by Trevor Hastie and Jonathan Taylor. His research interests are in statistics, machine learning, and optimization. Lately, he has worked on high dimensional statistical inference, analysis of non-convex optimization algorithms, and theory for deep learning.

+ Show Research List - Hide Research List
Xuanqing Liu, Cho-Jui Hsieh, Jason Lee, Yuekai Sun () "An inexact subsampled proximal Newton-type method for large-scale machine learning ,"  Submitted to Journal of Machine Learning Research.
Jason Lee () "Multiclass Clustering using a Semidefinite Relaxation ,"  Tech Report.
Mor Nacson, Jason Lee, Suriya Gunasekar, Nathan Srebro, Daniel Soudry () "Convergence of Gradient Descent on Separable Data ,"  Artificial Intelligence and Statistics (AISTATS).
Mahdi Soltanolkotabi, Adel Javanmard, Jason Lee () "Theoretical insights into the optimization landscape of over-parameterized shallow neural networks ,"  IEEE Trans. on Information Theory , DOI:10.1109/TIT.2018.2854560.
Michael Jordan, Jason Lee, Yun Yang () "Communication-efficient distributed statistical learning ,"  Journal of the American Statistics Association.
Chenwei Wu, Jiajun Luo, Jason Lee () "No Spurious Local Minima in a Two Node Neural Network ,"  International Conference on Learning Representations (ICLR) Workshop Track.
Simon Du, Jason Lee, Yuandong Tian () "When is a Convolutional Filter Easy to Learn? ,"  International Conference on Learning Representations (ICLR).
Maher Nouiehed, Jason Lee, Meisam Razaviyayn () "Convergence to Second-Order Stationarity for Constrained Non-Convex Optimization ,"  Submitted to SIAM Journal on Optimization.
Colin Wei, Jason Lee, Qiang Liu, Tengyu Ma () "On the Margin Theory of Feedforward Neural Networks ,"  arXiv preprint arXiv:1810.05369.
Jason Lee () "Please see personal website http://www-bcf.usc.edu/~lee715/ for up-to-date publication list ,".
Jason Lee, Ioannis Panageas, Georgios Piliouras, Max Simchowitz, Michael Jordan, Benjamin Recht () "First-order Methods Almost Always Avoid Saddle Points ,"  Accepted at Math Programming.
Simon Du, Jason Lee () "On the Power of Over-parametrization in Neural Networks with Quadratic Activation ,"  International Conference on Machine Learning (ICML).
Simon Du, Jason Lee, Haochuan Li, Liwei Wang, Xiyu Zhai () "Gradient Descent Finds Global Minima of Deep Neural Networks ,"  arXiv preprint arXiv:1811.03804.
Sham Kakade, Jason Lee () "Provably Correct Automatic Subdifferentiation for Qualified Programs ,"  Neural Information Processing Systems (NIPS).
Simon Du, Jason Lee, Yuandong Tian, Barnabas Poczos, Aarti Singh () "Gradient Descent Learns One-hidden-layer CNN: Don't be Afraid of Spurious Local Minima ,"  International Conference on Machine Learning (ICML).
Maziar Sanjabi, Jimmy Ba, Meisam Razaviyayn, Jason Lee () "Solving Approximate Wasserstein GANs to Stationarity ,"  Neural Information Processing Systems (NIPS).
Shiyu Liang, Ruoyu Sun, Jason Lee, R Srikant () "Adding One Neuron Can Eliminate All Bad Local Minima ,"  Neural Information Processing Systems (NIPS).
Mingyi Hong, Jason Lee, Meisam Razaviyayn () "Gradient Primal-Dual Algorithm Converges to Second-Order Stationary Solutions for Nonconvex Distributed Optimization ,"  International Conference on Machine Learning (ICML).
Xi Chen, Jason Lee, Xin Tong, Yichen Zhang () "Statistical Inference for Model Parameters in Stochastic Gradient Descent ,"  Accepted at Annals of Statistics.
Simon Du, Wei Hu, Jason Lee () "Algorithmic Regularization in Learning Deep Homogeneous Models: Layers are Automatically Balanced ,"  Neural Information Processing Systems (NIPS).
Suriya Gunasekar, Jason Lee, Daniel Soudry, Nathan Srebro () "Implicit Bias of Gradient Descent on Linear Convolutional Networks ,"  Neural Information Processing Systems (NIPS).
Jason Lee () "Stochastic Subgradient Converges in Polynomial Time on Nonsmooth Functions ,"  Unpublished.
Suriya Gunasekar, Jason Lee, Daniel Soudry, Nathan Srebro () "Characterizing Implicit Bias in Terms of Optimization Geometry ,"  International Conference on Machine Learning (ICML).
Rong Ge, Jason Lee, Tengyu Ma () "Learning One-hidden-layer Neural Networks with Landscape Design ,"  International Conference on Learning Representations (ICLR).
Damek Davis, Dmitriy Drusvyatskiy, Sham Kakade, Jason Lee () "Stochastic subgradient method converges on tame functions ,"  Foundations of Computational Mathematics.
Adel Javanmard, Jason Lee () "A Flexible Framework for Hypothesis Testing in High-dimensions ,".
Simon Du, Chi Jin, Jason Lee, Michael Jordan, Aarti Singh, Barnabas Poczos () "Gradient Descent Can Take Exponential Time to Escape Saddle Points ,"  Neural Information Processing Systems (NIPS).
Rong Ge, Jason Lee, Tengyu Ma () "Matrix Completion has no Spurious Local Minima ,"  Neural Information Processing Systems (NIPS).
Jason Lee, M. Simchowitz, M. Jordan, B. Recht () "Gradient Descent Converges to Minimizers. ,"  Conference on Learning Theory (COLT).
Jason Lee, Tengyu Ma, Qihang Lin, Tianbao Yang () "Distributed Stochastic Variance Reduced Gradient Methods ,"  Journal of Machine Learning Research.
Jason Lee, Dennis Sun, Yuekai Sun, Jonathan Taylor () "Exact Inference after Model Selection via the Lasso ,"  Annals of Statistics.
Jason Lee, Dennis Sun, Yuekai Sun, Jonathan Taylor () "Exact Post-Selection Inference with the Lasso. ,"  Annals of Statistics.
Yuchen Zhang, Jason Lee, Michael Jordan () "l1-regularized Neural Networks are Improperly Learnable in Polynomial Time ,"  International Conference on Machine Learning (ICML).
Qiang Liu, Jason Lee, Michael Jordan () "A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluation ,"  International Conference on Machine Learning (ICML).
Yuchen Zhang, Jason Lee, Martin Wainwright, Michael Jordan () "Learning Halfspaces and Neural Networks with Random Initialization ,"  Submitted..
Qiang Liu, Jason Lee () "Black-box importance sampling ,"  arXiv preprint arXiv:1610.05247.
Rong Ge, Jason Lee, Tengyu Ma () "Matrix Completion has No Spurious Local Minimum ,"  Neural Information Processing Systems (NIPS).
Jason Lee, Qiang Liu, Yuekai Sun, Jonathan Taylor () "Communication-Efficient Distributed Sparse Regression ,"  Journal of Machine Learning Research.
Jialei Wang, Jason Lee, Mehrdad Mahdavi, Mladen Kolar, Nathan Srebro () "Sketching Meets Random Projection in the Dual: A Provable Recovery Algorithm for Big and High-dimensional Data ,"  arXiv preprint arXiv:1610.03045.
Michael Jordan, Jason Lee, Yun Yang () "Communication-efficient distributed statistical learning ,"  arXiv preprint arXiv:1605.07689.
Jason Lee, Yuekai Sun, Jonathan Taylor () "Evaluating the Statistical Significance of Biclusters ,"  Neural Information Processing Systems (NIPS), 1–9.
Trevor Hastie, Rahul Mazumder, Jason Lee, Reza Zadeh () "Matrix Completion and Low-Rank SVD via Fast Alternating Least Squares ,"  Journal of Machine Learning Research.
Jason Lee, Yuekai Sun, Jonathan Taylor () "On Model Selection Consistency of Regularized M-Estimators ,"  Electronic Journal of Statistics.
Jason Lee, Jonathan Taylor () "Exact Post Model Selection Inference for Marginal Screening ,"  Neural Information Processing Systems (NIPS), 1–9.
Jason Lee, Trevor Hastie () "Learning the Structure of Mixed Graphical Models ,"  Journal of Computational and Graphical Statistics.
Jason Lee, Yuekai Sun, Michael Saunders () "Proximal Newton-Type Methods for Minimizing Composite Functions ,"  SIAM Journal on Optimization  24, 1420-1443.
Austin Benson, Jason Lee, Bartek Rajwa, David Gleich () "Scalable Methods for Nonnegative Matrix Factorizations of Near-Separable Tall-and-Skinny Matrices ,"  Neural Information Processing Systems (NIPS), 1–9.
Jason Lee, Trevor Hastie () "Structure Learning of Mixed Graphical Models ,"  Artificial Intelligence and Statistics (AISTATS), 388--396.
Jason Lee, Ran Gilad-Bachrach, Rich Caruana () "Using Multiple Samples to Learn Mixture Models ,"  Neural Information Processing Systems (NIPS), 324--332.
Jason Lee, Yuekai Sun, Jonathan Taylor () "On Model Selection Consistency of Penalized M-Estimators: a Geometric Theory ,"  Neural Information Processing Systems (NIPS), 342--350.
Jason Lee, Yuekai Sun, Michael Saunders () "Convergence Analysis of Inexact Proximal Newton-Type Methods ,"  NIPS Workshop on Optimization in Machine Learning.
Jason Lee, Yuekai Sun, Michael Saunders () "Proximal Newton-type Methods for Convex Optimization ,"  Neural Information Processing Systems (NIPS), 836--844.
Jason Lee, Mauro Maggioni () "Multiscale Analysis of Time Series of Graphs ,"  International Conference on Sampling Theory and Applications (SAMPTA).
Markus Kliegl, Jason Lee, Jun Li, Xinchao Zhang, Chuanxiong Guo, David Rinc\'on () "Generalized DCell Structure for Load-Balanced Data Center Networks ,"  IEEE Conference on Computer Communications (INFOCOM), 1--5.
Jason Lee () "Multiscale Estimation of Intrinsic Dimensionality of Point Cloud Data and Multiscale Analysis of Dynamic Graphs ,"  Senior Thesis, Duke University.
Jason Lee, Ben Recht, Nathan Srebro, Joel Tropp, Ruslan Salakhutdinov () "Practical Large-Scale Optimization for Max-Norm Regularization ,"  Neural Information Processing Systems (NIPS), 1297--1305.
Anna Little, Jason Lee, Yoon-Mo Jung, Mauro Maggioni () "Estimation of Intrinsic Dimensionality of Samples from Noisy Low-Dimensional Manifolds in High Dimensions with Multiscale SVD ,"  IEEE Workshop on Statistical Signal Processing (SSP), 85--88.
Markus Kliegl, Jason Lee, Jun Li, Xinchao Zhang, Chuanxiong Guo, David Rinc\'on () "Generalized DCell Structure for Load-Balanced Data Center Networks ,"  Microsoft Research Technical Report, 1--14.
Jason Lee, John Neuberger () "Existence of Asymptotic Solutions to Semi-linear Partial Difference Equations ,"  Joint Mathematics Meetings.