Department of Electrical Engineering and Computer Science
Office: CST 3-127/3-122
Dr. Yingbin Liang received the Ph.D. degree in Electrical Engineering from the University of Illinois at Urbana-Champaign in 2005. In 2005-2007, she was working as a postdoctoral research associate at Princeton University. In 2008-2009, she was an assistant professor at the Department of Electrical Engineering at the University of Hawaii. Since December 2009, she has been a faculty member at the Department of Electrical Engineering and Computer Science at the Syracuse University. Dr. Liang's research interests include machine learning, statistical learning theory, high dimensional data analysis, parallel and distributed optimization, information theory, and wireless networks.
My research spans over multiple disciplines including statistical signal processing, machine learning, large-scale optimization, information theory, and wireless communications and networks. More specifically, my research projects can be summarized along the following three main themes. .1. Developing innovative machine learning and signal processing methodologies and algorithms that exploit sparsity and statistical structures in data for handling high-dimensional large-scale data with provable performance guarantee. Specific projects include multi-task high-dimensional linear regression and applications in genomic data analysis; robust principle component analysis (PCA) under nonuniform sparse errors and applications to community detection; minimax optimal estimation of KL divergence between large-alphabet distributions; adaptive sampling and signal demixing in compressed sensing; kernel-based detection of anomalous events in networks; and nonparametric decentralized detection via sparse learning. 2. Developing fast optimization algorithms for large-scale machine learning applications that involve ultra-high dimensional data and understanding the performance guarantee of these algorithms. Specific projects include distributed asynchronous parallel algorithms for nonconvex nonsmooth functions, stochastic model-parallel algorithms for online data processing, nonconvex methods and fast algorithms for phase retrieval, stochastic algorithm and its convergence for phase retrieval, a unified view of proximal algorithm via general distance metric. 3. Developing novel technologies for data transmission over various challenging communication media and different networked scenarios, and analyzing the impact of these technologies on the fundamental performance limits of data communications subject to a number of requirements including error probability, secrecy, and delay. Specific projects include layered secure coding for broadcast networks, secret multiple key generation over various networks and key capacity region, fundamental rate limits of state-dependent networks and the impact on interference management in wireless communications, and communication over time-varying and MIMO Poisson channels and networks.
Selected Publications by TopicsFull Journal Publications and Book Chapters Full Conference Publications