Tengyuan Liang

Assistant Professor of Econometrics and Statistics

Biography

Dr. Liang is an assistant professor at Chicago Booth. He is also the George C. Tiao faculty fellow in data science research.

His current research interests include studying the computational and algorithmic aspects of statistical inference, reconciling the empirical and theoretical gap in modern machine learning and statistical learning, and exploiting stochastic methods in non-convex optimization.

Interests

Data Science
Statistical Inference
Statistical Learning
Stochastic Optimization
Economic and Business Applications

Education

PhD in Statistics, 2017

Wharton School at University of Pennsylvania
BSc in Mathematics and Applied Mathematics, 2012

Peking University

Manuscripts

Tengyuan Liang, Pragya Sur. A Precise High-Dimensional Asymptotic Theory for Boosting and Min-L1-Norm Interpolated Classifiers. arXiv:2002.01586, 2020.

Preprint PDF Project Project Project

Tengyuan Liang. Estimating Certain Integral Probability Metrics (IPMs) Is as Hard as Estimating under the IPMs. arXiv:1911.00730, 2019.

Preprint PDF Project Project

Tengyuan Liang, Alexander Rakhlin, Xiyu Zhai. On the Multiple Descent of Minimum-Norm Interpolants and Restricted Lower Isometry of Kernels. arXiv:1908.10292, 2019.

Preprint PDF Project Project Project

Tengyuan Liang. On How Well Generative Adversarial Networks Learn Densities: Nonparametric and Parametric Results. arXiv:1811.03179, 2019.

Preprint PDF Project Project Project

Max Farrell, Tengyuan Liang, Sanjog Misra. Deep Neural Networks for Estimation and Inference: Application to Causal Effects and Other Semiparametric Estimands. arXiv:1809.09953, 2019.

Preprint Project Project Project Project

Publications

Xialiang Dou, Tengyuan Liang. Training Neural Networks as Learning Data-adaptive Kernels: Provable Representation and Approximation Benefits. Journal of the American Statistical Association (Theory and Methods), to appear, 2020.

Preprint PDF Project Project Project

T. Tony Cai, Tengyuan Liang, Alexander Rakhlin. Weighted Message Passing and Minimum Energy Flow for Heterogeneous Stochastic Block Models with Side Information. Journal of Machine Learning Research, 2020.

Preprint PDF Project

Tengyuan Liang, Alexander Rakhlin. Just Interpolate: Kernel ''Ridgeless'' Regression Can Generalize. Annals of Statistics, to appear, 2019.

Preprint PDF Project Project Project

Tengyuan Liang, Weijie Su. Statistical Inference for the Population Landscape via Moment Adjusted Stochastic Gradients. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 2019.

Preprint PDF Project Project

Tengyuan Liang, James Stokes. Interaction Matters: A Note on Non-asymptotic Local Convergence of Generative Adversarial Networks. International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.

Preprint PDF Project Project

Tengyuan Liang, Tomaso Poggio, Alexander Rakhlin, James Stokes. Fisher-Rao Metric, Geometry, and Complexity of Neural Networks. International Conference on Artificial Intelligence and Statistics (AISTATS), 2019.

Preprint PDF Project Project Blog

Belinda Tzen, Tengyuan Liang, Maxim Raginsky. Local Optimality and Generalization Guarantees for the Langevin Algorithm via Empirical Metastability. Conference on Learning Theory (COLT), 2018.

Preprint PDF Project Project

Satyen Kale, Zohar Karnin, Tengyuan Liang, Dávid Pál. Adaptive Feature Selection: Computationally Efficient Online Sparse Linear Regression under RIP. International Conference on Machine Learning (ICML), 2017.

Preprint PDF Project Project

T. Tony Cai, Tengyuan Liang, Alexander Rakhlin. Computational and Statistical Boundaries for Submatrix Localization in a Large Noisy Matrix. Annals of Statistics, 2017.

Preprint PDF Project

T. Tony Cai, Tengyuan Liang, Alexander Rakhlin. On Detection and Structural Reconstruction of Small-World Random Networks. IEEE Transactions on Network Science and Engineering, 2017.

Preprint PDF Project

T. Tony Cai, Tengyuan Liang, Alexander Rakhlin. Geometric Inference for General High-Dimensional Linear Inverse Problems. Annals of Statistics, 2016.

Preprint PDF Project

Tengyuan Liang, Alexander Rakhlin, Karthik Sridharan. Learning with Square Loss: Localization through Offset Rademacher Complexity. Conference on Learning Theory (COLT), 2015.

Preprint PDF Project

Alexandre Belloni, Tengyuan Liang, Hariharan Narayanan, Alexander Rakhlin. Escaping the Local Minima via Simulated Annealing: Optimization of Approximately Convex Functions. Conference on Learning Theory (COLT), 2015.

Preprint PDF Project

T. Tony Cai, Tengyuan Liang, Harrison H. Zhou. Law of Log Determinant of Sample Covariance Matrix and Optimal Estimation of Differential Entropy for High-Dimensional Gaussian Distributions. Journal of Multivariate Analysis, 2015.

Preprint PDF Project

Workshops & Talks

More Talks

Duke

Mar 25, 2020 12:00 AM

TRIPODS workshop on Theory and Modeling of Deep Learning

MIT

Jan 27, 2020 12:00 AM

MIFODS workshop on Learning with a Complex Structure

Harvard

Sep 16, 2019 12:00 AM

Statistics Colloquium

Yale

Sep 11, 2019 12:00 AM

Econometrics Seminar, Dept. of Economics

MIT

Sep 6, 2019 12:00 AM

Statistics and Stochastics Seminar Series

Teaching

Booth:

41000 (MBA): Business Statistics: Fall 19, Fall 18, Spring 18
Data Science Reading Group

Wharton:

622 (MBA): Advanced Quantitative Modeling: Spring 15, Spring 14

Awards

George C. Tiao Faculty Fellowship, 2017-2018
for research in computational and data science
J. Parker Memorial Bursk Award, 2016
for excellence in research
US Junior Oberwolfach Fellow, 2015
Winkelman Fellowship, 2014-2017
the highest honorific fellowship awarded by the Wharton School