Shirin Jalali

Assistant Professor


Phone:(848) 445-5309
Office:CoRE 721

Shirin Jalali is an Assistant Professor of Electrical and Computer Engineering at Rutgers University. Prior to joining Rutgers, she was a Research Scientist at Nokia Bell Labs. Her main background is in information theory, statistical signal processing and machine learning. Currently, her research interests span a range of problems related to i) developing theoretically-founded computationally-efficient  solutions to understand and address various issues rising in modern computational imaging, such as speckle noise and snapshot imaging, ii) online learning, iii) structure learning, and iv) developing an information theoretic understanding of neural networks and deep learning.


Ph.D. in Electrical Engineering from Stanford University 2010
M.Sc. in Statistics from Stanford University 2010
M.Sc. in Electrical Engineering  from Sharif University of Technology, 2004
B.Sc. in Electrical Engineering from Sharif University of Technology, 2002

Research Interests

  • High-dimensional inference and inverse problems 
  • Computational imaging
  • Machine learning
  • Information theory
  • Statistical signal processing 

Selected Publications

M. Bakhshizade, A. Maleki, S. Jalali, "Using black-box compression algorithms for phase retrieval", IEEE Transactions on Information Theory, vol. 66, no. 12., p.p. 7978-8001, Dec. 2020. 

P. Peng, S. Jalali and X. Yuan, "Solving inverse problems via auto-encoders",  IEEE Journal on Selected Areas in Information Theory, vol. 1, no. 1, pages 312-323, May 2020 [arxiv] [code]

S. Jalali,  "Toward theoretically-founded learning-based compressed sensing", IEEE Transactions on Information Theory, vol. 66 , no. 1, pages 387-400, Jan. 2020

S. Jalali, X. Yuan,, "Snapshot compressed sensing: performance bounds and algorithms", IEEE Transactions on Information Theory,vol. 65, no. 12, pages 8005 - 8024 Dec. 2019 [arxiv]

S. Jalali, C. Nuzman, I. Saniee, "Efficient deep approximation of GMMs", Conference on Neural Information Processing Systems (NeurIPS), Vancouver, Canada, 2019 

S. Beygi, S. Jalali, A. Maleki, U. Mitra, "An efficient algorithm for compression-based compressed sensing", Information and Inference: A Journal of IMA, Aug. 2018 

 S. Jalali, H. V. Poor, "Universal compressed sensing for almost lossless recovery", IEEE Transactions on Information Theory, vol. 63, no. 5, pp.  2933 - 2953, May 2017
S. Jalali, and A. Maleki, "From compression to compressed sensing", Applied and Computational Harmonic Analysis (ACHA), vol. 40 (2), pp. 352-385, March 2016