Tutorials and Lecture Series

  1. Reinforcement Learning: Fundamentals, Algorithms, and Theory

    1. ICASSP 2022, together with Yuting Wei and Yuejie Chi

  2. Statistical and algorithmic foundations of reinforcement learning

    1. ICSA Applied Statistics Symposium 2021, together with Yuting Wei, Yuejie Chi and Zhengyuan Zhou

  3. Taming nonconvexity in information science

    1. ITW 2018, together with Yuejie Chi

  4. Recent advances in nonconvex methods for high-dimensional estimation

    1. ICASSP 2018, together with Yuejie Chi and Yue Lu

  5. TRIAD Lecture Series 2019, Georgia Tech

    1. The power of nonconvex optimization in solving random quadratic systems of equations

    2. Random initialization and implicit regularization in nonconvex statistical estimation

    3. Projected power method: an efficient nonconvex algorithm for joint discrete assignment

    4. Spectral methods meets asymmetry: two recent stories

    5. Inference and uncertainty quantification for noisy matrix completion

Invited Talks

  1. Inference and uncertainty quantification for low-rank models

    1. Statistics Seminar, UIUC, Oct. 2021

    2. Data Science Seminar, JHU, Oct. 2021

  2. Demystifying the efficiency of reinforcement learning: a few recent stories

    1. IEOR/DRO Seminar, Columbia University, Sep. 2021

    2. ISL Colloquium, Electrical Engineering, Stanford University, Apr. 2021

    3. ESE Seminar Series, University of Pennsylvania, May 2021

    4. Frontiers in Computing and Mathematical Sciences, Caltech, Feb. 2021

    5. Operation Research Seminar, Chinese Operation Research Society, Sep. 2020

  3. On the effectiveness of nonconvex optimization in reinforcement learning

    1. Semiautonomous Seminar, University of California, Berkeley, Oct. 2021

    2. ICERM Workshop on Advances in Theory and Algorithms for Deep Reinforcement Learning, Aug. 2021

  4. Taming nonconvexity in tensor completion: Fast convergence and uncertainty quantification

    1. IPAM Workshop on Efficient Tensor Representations for Learning and Computational Complexity, May 2021.

    2. Department Seminar, Statistics Department, Harvard University, Feb. 2021

    3. Department Seminar, Department of Data Sciences and Operations, University of Southern California, Feb. 2021

  5. Taming nonconvexity in statistical and reinforcement learning

    1. Department Seminar, Department of Electrical Engineering, Yale University, Feb. 2021

    2. ECSE Seminar Series, RPI, Feb. 2021

  6. Breaking the sample size barrier in reinforcement learning via model-based approaches (a.k.a. plug-in approaches)

    1. Neyman Seminar, Statistics Department, UC Berkeley, Nov. 2020

  7. Nonconvex optimization meets statistics: a few recent stories

    1. Department Seminar, Electrical and Computer Engineering, Carnegie Mellon University, Jan. 2021

    2. Communications and Signal Processing seminar, University of Michigan, March 2020

    3. TBSI Workshop on Data Science, Shenzhen, December 2019

    4. Penn Research in Machine Learning Seminar, University of Pennsylvania, Oct. 2019

    5. ORIE Colloquium Seminar Series, Cornell University, Oct. 2019

  8. Inference and uncertainty quantification for noisy matrix completion

    1. ICSA International Conference, Hangzhou, December 2019

  9. Bridging convex and nonconvex optimization in noisy matrix completion: stability and uncertainty quantification

    1. MIFODS Workshop on Graphical Models, MIT, August 2019

    2. International Conference on Continuous Optimization (ICCOPT), Berlin, August 2019

    3. Statistics Seminar, Stanford University, May 2019

    4. Statistics Colloquium, University of Chicago, May 2019

    5. PACM Colloquium, Princeton University, May 2019

  10. Stability, nonconvex optimization, and asymmetry in low-rank matrix estimation

    1. TRIPODS Seminar, University of Arizona, Feb. 2019

  11. Noisy matrix completion: understanding statistical guarantees of convex relaxation via nonconvex optimization

    1. Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, Nov. 2019

    2. INFORMS Annual Meeting, Seattle, Oct. 2019

    3. Information Theory and Its Applications (ITA) Workshop, San Diego, Feb. 2019

  12. Random initialization and implicit regularization in nonconvex statistical estimation

    1. TRIPODS Southwest Summer Conference, May 2019

    2. Signal and Information Processing (SIP) Seminar, Rutgers University, Mar. 2019

    3. KITP Conference ‘‘Rough Landscapes: From Physics to Algorithms’’, UCSB, Jan. 2019

    4. International Workshop on Mathematical Issues in Information Sciences (MIIS), Shenzhen, Dec. 2018

    5. MaD (Math and Data) Seminar, NYU Courant Institute and Center for Data Science, Nov. 2018

    6. INFORMS Annual Meeting, Phoenix, Nov. 2018

    7. Optimization Seminar, ORFE, Princeton University, Oct. 2018

    8. Department Seminar, Department of Statistics and Data Science, Yale University, Oct. 2018

    9. 56th Annual Allerton Conference on Communication, Control, and Computing, Monticello, Oct. 2018

    10. ISE Department Seminar, University of Southern California, Sep. 2018

    11. London Workshop on Non-convex Optimisation and Matrix Factorisation, Sep. 2018

    12. Wharton Statistics Department Seminar, University of Pennsylvania, Sep. 2018

    13. Statistics Research Colloquium, Purdue University, Sep. 2018

    14. ICSA China Conference with the Focus on Data Science, Qingdao, July 2018

    15. ICSA 2018 Applied Statistics Symposium, New Brunswick, June 2018

    16. Joint ISL Colloquium and IT Forum, Stanford University, May 2018

    17. Oberwolfach Workshop on Applied Harmonic Analysis and Data Processing, Mar. 2018

    18. Colloquium, Statistics Department, PSU, Apr. 2018

  13. Asymmetry helps: Eigenvalue and eigenvector analyses of asymmetrically perturbed low-rank matrices

    1. CMStatistics, Pisa, Dec. 2018

  14. Implicit regularization in nonconvex statistical estimation

    1. INFORMS Annual Meeting, Phoenix, Oct. 2019

    2. Information Theory and Its Applications (ITA) Workshop, San Diego, Feb. 2018

    3. Data Science Seminar series, Institute for Mathematics and its Applications (IMA), Jan. 2018

    4. International Conference on Data Science, Shanghai, Dec. 2017

    5. Signal Processing and Communications Seminar Series, University of Delaware, Dec. 2017

    6. Simons Institute Workshop on Optimization, Statistics and Uncertainty, Berkeley, Nov. 2017

    7. 50th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, Oct. 2017

    8. Statistics Seminar, Columbia University, Oct. 2017

  15. Regularized mirror descent: A nonconvex approach for learning mixed probability distributions

    1. 55th Annual Allerton Conference on Communication, Control, and Computing, Monticello, Oct. 2017

  16. Spectral method and regularized MLE are both optimal for top-K ranking

    1. Joint Statistical Meetings (JSM), Baltimore, Aug. 2017

  17. The projected power method: a nonconvex algorithm for discrete problems

    1. Electrical Engineering Seminar Series, Harvard University, Apr. 2017

  18. ‘‘The effectiveness of nonconvex optimization in two problems’’

    1. Statistics Seminar, NYU Stern School of Business, New York, Mar. 2017

    2. IDeAS Seminar, Princeton University, Mar. 2017

  19. The projected power method: an efficient algorithm for joint alignment from pairwise differences

    1. Information Theory and Applications (ITA) Workshop, San Diego, Feb. 2017

    2. Meeting of the International Linear Algebra Society, July 2017

    3. SIAM Conference on Optimization, Vancouver, May 2017

    4. Stanford Wireless Systems Lab, Stanford, Jan. 2017

    5. CMO-BIRS Workshop: Applied Harmonic Analysis, Massive Data Sets, Machine Learning, and Signal Processing, Oaxaca, Oct. 2016

    6. 54th Annual Allerton Conference on Communication, Control, and Computing, Monticello, Sep. 2016

  20. Solving random quadratic systems of equations is nearly as easy as solving linear systems

    1. ShanghaiTech Symposium on Information Science and Technology (SSIST) 2017, Shanghai, July 2017

    2. International Conference on Continuous Optimization (ICCOPT), Berlin, August 2019

    3. 50th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, Nov. 2016

    4. World Congress in Probability and Statistics, Toronto, July 2016

    5. Workshop on Sensing and Analysis of High-Dimensional Data (SAHD), Duke University, July 2015

  21. The power of nonconvex paradigms for high-dimensional estimation
    MIT EECS, Princeton EE/CS, USC CS/ISE, CMU MLD, UMich ECE/MIDAS, Columbia EE/IEOR, Cornell ECE, UC Davis ECE, NYU ECE, Jan. 2016 – Mar. 2016

  22. Modern optimization meets physics: recent progress on phase retrieval

    1. International Matheon Conference on Compressed Sensing and its Applications (CSA), Berlin, Dec. 2015

  23. Near-optimal joint object matching via convex relaxation

    1. INFORMS Annual Meeting, Phoenix, Nov. 2018

    2. IDeAS Seminar, Princeton University, Apr. 2014

    3. Information Initiative at Duke (iiD) Seminar, Duke University, Apr. 2014

    4. Center for Signal and Information Processing (CSIP) Seminar, Georgia Tech, Mar. 2014