Research

My research interests lie in areas of machine learning, statistics, and signal processing. I'm presently particularly interested in the following three projects:

Deep neural networks for inverse problems

Deep neural networks have become highly effective tools for compressing images and solving inverse problems including denoising, inpainting, and reconstruction of images from few and noisy measurements. I'm interested in developing corresponding algorithms and theory.

DNA data storage 

Active learning and related problems

Active learning uses data collected in the past to inform which data to collect in the future. This can often significantly reduce the sample complexity or the performance of an estimator.

DNA data storage

How can information efficiently be reconstructed from millions of noisy sequences? That question lies at the heart of an emerging technology that stores digital information long-term on DNA. Practical constraints on reading and writing DNA require the data to be stored on many short molecules, and imperfections in sequencing, synthesis, and handling, as well as decay of DNA lead to a loss of sequences and induce errors within the sequences. I'm interested in designing new algorithms and coding schemes for recovering information in DNA storage systems.

DNA data storage