Gang Wang
 |
Hello! I'm currently a Postdoctoral Associate in the Dept. of Electrical and Computer Engineering at the Univ. of Minnesota, working with Prof. Georgios B. Giannakis, where I obtained a Ph.D. degree in 2018. I have also earned a Ph.D. degree in Control Science and Engineering under the supervision of Prof. Jie Chen, and a B.Eng. degree in Electrical Engineering and Automation, both from the Beijing Institute of Technology, Beijing, China. |
Contact information
Digital Technology Center
University of Minnesota, Twin Cities
117 Pleasant St SE, Minneapolis, MN, 55455
Office: Room 427 Walter Library
E-mail: gangwang AT umn.edu.
Recent updates
1/2020: Two papers were accepted to The 45th International Conference on Acoustics, Speech, and Signal Processing (ICASSP’2020).
1/2020: Two papers were accepted to The 23rd International Conference on Artificial Intelligence and Statistics (AISTATS’2020).
‘‘Finite-Time Error Bounds for Biased Stochastic Approximation Algorithms with Application to Q-Learning’’ This paper develops a multistep Lyapunov approach for non-asymptotic analysis of biased stochastic approximation. When applying our results to TD- and Q-learning with (non)linear function approximation, we provide the first finite-time error bounds for TD- and Q-learning that hold from the first iteration and Markov chains starting with any initial distribution.
‘‘Finite-Time Analysis of Decentralized Temporal-Difference Learning with Linear Function Approximation’’ This paper develops the first finite-time error bound for fully decentralized TD-learning under Markovian data samples.
12/2019: Paper ‘‘Wireless Power Transmitter Deployment Balancing Fairness and Charging Service Quality" was accepted to the IEEE Internet of Things Journal.
11/2019: Received the Excellent Doctoral Dissertation Award from the Chinese Association of Automation (in Chinese). This award recognizes excellent thesis research by doctoral candidates in the field of automation in China; presented annually to up to ten authors.
11/2019: Served as a TPC member of the 2019 IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP).
10/2019: Paper “Learning Connectivity and Higher-Order Interactions in Radial Distribution Grids” submitted to the 2020 International Conference on Acoustics, Speech, and Signal Processing (pdf).
10/2019: Check our paper “Finite-Time Analysis of Decentralized Temporal-Difference Learning with Linear Function Approximation”, which provides finite-time error bounds for decentralized temporal-difference learning from Markovian data samples (pdf).
10/2019: Paper ‘‘Two-Timescale Voltage Control in Distribution Grids Using Deep Reinforcement Learning’’ was accepted to the IEEE Transactions on Smart Grid.
9/2019: Paper ‘‘Resonant Beam Communications: Principles and Designs" was accepted to the IEEE Communications Magazine.
9/2019: Paper ‘‘Optimal Switching Data Injection Attacks and Countermeasures in Cyber-Physical Systems” was accepted to the IEEE Transactions on Systems, Man, and Cybernetics: Systems.
9/2019: Appointed as editorial board member of the Signal Processing journal.
9/2019: My first paper on reinforcement learning theory “A Multistep Lyapunov Approach for Finite-Time Analysis of Biased Stochastic Approximation” (pdf).
8/2019: Presented our paper “Fast LAV Estimation via Composite Optimization” in the Best Conference Papers session at the 2019 IEEE Power and Energy Society General Meeting.
7/2019: Paper “Deep Reinforcement Learning for Adaptive Caching in Hierarchical Content Delivery Networks” was accepted to the IEEE Transactions on Cognitive Communications and Networking.
6/2019: Served as a TPC member of the *International Conference on Wireless Communications and Signal Processing/.
6/2019: Paper “Mobile Energy Transfer in Internet of Things” was accepted to the IEEE Internet of Things Journal.
5/2019: Paper “Real-Time Power System State Estimation and Forecasting via Deep Unrolled Neural Networks” was accepted to the IEEE Transactions on Signal Processing.
2/2019: Served as a TPC member of the IEEE Data Science Workshop.
1/2019: Paper “Robust and Scalable Power System State Estimation via Composite Optimization” was accepted to the IEEE Transactions on Smart Grid.
11/2018: Paper on nonlinear discriminative dimensionality reduction was presented in Asilomar.
10/2018: Paper “Robust Power System State Estimation from Rank-One Measurements” was accepted to the IEEE Transactions on Control of Network Systems.
8/2018: My first paper on deep learning theory “Learning Single-Hidden-Layer ReLU Networks on Linearly Separable Data: Algorithms, Optimality, and Generalization” (Website and codes).
6/2018: Paper on graph canonical correlation analysis was accepted to the IEEE Transactions on Signal Processing.
3/2018: Defended thesis on ‘‘Non-convex Phase Retrieval Algorithms and Performance Analysis.”
3/2018: Check our new submission on graph canonical correlation analysis. Check it out here.
3/2018: Paper ‘‘Phase Retrieval via Reweighted Amplitude Flow” was accepted to the IEEE Transactions on Signal Processing. Matlab codes available for download here.
10/2017: Paper on ‘‘Sparse Phase Retrieval via Truncated Amplitude Flow” accepted to the IEEE Transactions on Signal Processing. Matlab codes available for download (Website and codes).
10/2017: Paper on ‘‘Discriminative Principal Component Analysis” for joint analysis of multiple large-scale datasets submitted. Available here. Check it out!
9/2017: We won a Best Student Paper Award in the 2017 European Conference on Signal Processing (EUSIPCO’2017) for the joint work with Profs. G. B. Giannakis and J. Chen on efficient algorithms for large-scale phase retrieval, which was selected from several hundred submissions based on rigorous evaluations of the selection committee.
8/2017: Paper coauthored with Profs. G. B. Giannakis, Y. Saad (CS, UMN), and J. Chen (BIT) accepted to NIPS’2017 (Acceptance rate: 21%), which presents a new algorithm for solving systems of random quadratic equations benchmarking the numerical performance. Matlab codes available for download here.
8/2017: Paper on randomized block Franke-Wolfe algorithms, joint work with L. Zhang, D. Romero (Univ. of Agder), accepted to the IEEE Transactions on Signal Processing, in which we develop a rich family of feasible step sizes for running Frank-Wolfe in parallel. Congrats Liang!
8/2017: Paper on scalable composite optimization algorithms for robust LAV (least-absolute-value) power system state estimation submitted. Check it out in Publications.
7/2017: Paper “Solving Random Systems of Quadratic Equations via Truncated Amplitude Flow,” joint work with Prof. Y. Eldar (Technion), was accepted to the IEEE Transactions on Information Theory. Matlab codes available (Website and codes).
5/2017: Invited book chapter on “Advances in Power System State Estimation,” coauthored with Profs. V. Kakatos (VT), H. Zhu (UT Austin), and G. B. Giannakis submitted. The book chapter offers a contemporary view of state estimation for modern autonomous energy grids.
1/2017: Paper “Solving Large-scale Random Systems of Quadratic Equations via Stochastic Truncated Amplitude Flow” accepted to the IEEE Transactions on Signal Processing. Matlab codes available (Website and codes).
12/2016: Paper “Solving Random Systems of Quadratic Equations via Truncated Generalized Gradient Flow” was presented at the NIPS’2016. A full version along with Matlab implementations is available at (Website and codes).