Sparse Inversion of Mixture and Bilinear Models via Convex Optimization

ECE Seminar: Sparse Inversion of Mixture and Bilinear Models via Convex Optimization

Starts at: October 8, 2015 4:30 PM

Ends at: 6:00 PM

Location: Scaife 125

Speaker: Dr. Yuejie Chi

Affiliation: Ohio State University

Refreshments provided: Yes

Link to Poster

Link to Video (1)


In many applications of engineering and applied science, the observation can be regarded as passing a point source signal through some point spread function, and the goal is to invert the locations and amplitudes of the point source signal from the observation. This problem has been well studied in the signal processing literature, resulting in many algorithms such as Prony, MUSIC, or ESPRIT.
In this talk, we consider two variations of the above standard problem where conventional spectrum estimation approaches no longer apply. The first variation is a mixture model, where one wishes to simultaneously identify the membership and locations and amplitudes of point sources that pass through different point spread functions, from their superpositions. This problem is motivated by three-dimensional super-resolution microscopy imaging and neural spike sorting. The second variation is a bilinear model, where one wishes to recover both the point source signal and the point spread function, when the latter is assumed unknown but mildly constrained in a low-dimensional subspace. This problem is motivated by blind channel estimation and self-calibration of sensor arrays. We develop convex optimization algorithms that are computationally efficient based on the atomic norm for spectrally-sparse signals, and establish their performance guarantees. Numerical examples will be given to demonstrate the effectiveness of the proposed approaches. This is joint work with my student Yuanxin Li.

Dr. Yuejie Chi is an assistant professor in the Electrical and Computer Engineering Department at The Ohio State University, with a joint appointment in the Biomedical Informatics Department at the Wexner Medical School since September 2012.
She received a M.A. and a Ph.D. in Electrical Engineering from Princeton University in 2009 and 2012 respectively, and a B.Eng. in Electrical Engineering from Tsinghua University, China in 2007.
She received the Young Investigator Program Awards from AFOSR and ONR respectively in 2015, and the Ralph E. Powe Junior Faculty Enhancement Award from ORAU in 2014. She is the recipient of the IEEE Signal Processing Society Young Author Best Paper Award in 2013 and the Best Paper Award at the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) in 2012.
Her research interests include high-dimensional data analysis, statistical signal processing, machine learning and their applications in network inference, spectrum sensing and estimation, image analysis and bioinformatics.