Junan Zhu (朱俊安)

Ph.D.

Department of Electrical and Computer Engineering

North Carolina State University

Contact: jzhu9@ncsu.edu

Education Background


Ph.D. 2011-2016 Department of Electrical and Computer Engineering North Carolina State University, U.S. Advisor: Dror Baron
B.E. 2007-2011 School of Optical Electrical and Computer Engineering University of Shanghai for Science and Technology, China. Advisor: Yiming Zhu

News


2016.12.2 Junan defended his Ph.D. Cheers! (Dissertation pdf,Dissertation in IEEE style bib.)
2016.11.7 Junan submitted the paper "Optimal trade-offs in multi-processor approximate message passing" with A. Beirami and D. Baron (ArXiv,pdf).
2016.7.26 Junan's joint work with Y. Ma and D. Baron "Approximate message passing algorithm with universal denoising and Gaussian mixture learning" got accepted to IEEE Trans. Signal Process.(ArXiv,pdf,IEEE version).
2016.4.8 Junan submitted the work with D. Baron and F. Krzakala, "Performance limits for noisy multi-measurement vector problems" (ArXiv,pdf).
2016.4.3 Junan's paper with A. Beirami and D. Baron, "Performance trade-offs in multi-processor approximate message passing," is accepted into ISIT2016 (ArXiv,pdf).
2016.3.1 Junan passed his Prelim exam. Cheers!
2016.2.4 D. Baron presented "Large-scale multi-processor approximate message passing with lossy compression" at ITA (slides).
2016.1.17 Junan submitted "Multi-processor approximate message passing with lossy compression" (arxiv,pdf).
2015.7.9 Y. Ma made a short tutorial video about the approximate message passing with universal denoiser algorithm presented in our works with D. Baron (video link).
2015.7.7 Junan made a short tutorial video about the size- and level- adaptive Markov chain Monte Carlo algorithm presented in the works with M. F. Duarte and D. Baron (video link).
2015.6.8 Junan's joint work with Y. Ma and D. Baron "Approximate message passing with universal denoising" was submitted (arxiv,pdf,tutorial video).
2015.2.5 The work "Universal denoising and approximate message passing" with Y. Ma and D. Baron was presented at Information Theory Applications Workshop, San Diego, CA, February 2015 (talk).
2014.12.30 The paper "Complexity-matching universal signal estimation for compressed sensing" got accepted into the IEEE Transactions on Signal Processing with the new name "Recovery from linear measurements with complexity-matching universal signal estimation," and it is available online (IEEE version, arxiv,tutorial about the presented algorithm).
2014.8.13 The paper "Compressed sensing via universal denoising and approximate message passing" got accepted into the 52nd Allerton Conference on Communication, Control, and Computing, Monticello, IL, October 2014 (arxiv, pdf, slides).
2014.7.9 Junan submitted "Complexity-matching universal signal estimation for compressed sensing" (arxiv).
2014.7.7 Junan submitted "Compressed sensing via universal denoising and approximate message passing" (arxiv,pdf)
2014.4.20 Junan's paper "Complexity-adaptive universal signal estimation for compressed sensing" got accepted by SSP (IEEE version, pdf,poster).
2013.3.20 Junan attended CISS at Johns Hopkins University and presented his paper "Performance regions in compressed sensing from noisy measurements" (arxiv, final version, slides, IEEE version)
2013.2.8 Junan passed his Qualifying Review Exam. Cheers!
2013.2.5 Junan's paper "Performance regions in compressed sensing from noisy measurements" has been accepted and will appear in CISS 2013 (arxiv).
2013.1.23 We have made a video advertising the talk by Dr. Baron and J. Tan at ITA workshop on Feb. 14, 2013.
2013.1.4 Junan submitted his first conference paper "Performance regions in compressed sensing from noisy measurements" to CISS.
2011.8.17 Junan becomes a new graduate student in ECE at NCSU. He is now in Dr. Baron's group.
2011.7.29 Junan's paper "Theoretical study of W-shaped optical fiber with a depression in core center by applying analytical transfer matrix method" is now available online. It is part of the research work that he has done in his undergraduate studies.

Research Interests


In 2011, Junan became a graduate student at N.C. State University. He is interested in statistical modeling and statistical signal processing. In particular, Junan's research focuses on designing linear model learning algorithms, studying the optimal estimation quality for linear model learning algorithms, and exploring the optimal trade-offs among different costs for distributed learning algorithms. Prior to his graduate education, Junan focused on optics in his undergraduate studies at USST. To be more specific, he has done some research on the terahertz waveguide and optical fiber. Later, in his graduation thesis he carried out experimental research on fabrication of microstructures on silicon surface.

Papers


       

 

 

 

 

Nocost
Free Counter