I am a Ph.D. candidate in the Department of Electrical Engineering at UCLA. I am part of the LORIS Laboratory and am fortunate to have Prof. Lara Dolecek as an advisor.
Before joining UCLA, I studied Electrical Engineering at the University of Michigan, Ann Arbor.
My academic CV and my industry resume.
My interests are broadly related to the reliability of data. I am interested in questions about how to reliably store data on unreliable hardware, how to protect data from noise and uncertainty, how to reconstruct/reconcile data that has already been modified or corrupted, and how to design algorithms that can tolerate unreliable input data.
My approach to these problems involves tools from many disciplines: information theory, error-correcting codes, statistical learning, algorithms, combinatorics and optimization. Projects I have worked on include:
Data reconstruction and noisy statistical learning: Many algorithms focus on finding some core, original data from a number of heavily modified descendant files that contain the core data (finding an ancestral DNA sequence from a variety of modern genes, text reconstruction algorithms, etc…). My work seeks to answer how many descendant files are needed to perform the reconstruction correctly given certain structural conditions in the files (redundancy, maximum rate of change, etc…). I am also interested in how to make statistical learning algorithms robust to noisy, unreliable data: if we have a limited protection budget, how should we choose to protect features to limit divergence from the idealized, noiseless result?
Efficient data synchronization: Efficiently reconciling or synchronizing copies of files that differ (possibly only by a little) is a crucial problem in cloud and backup storage. Our work proposes a novel file synchronization algorithm that works well in theory (nearly optimal under certain conditions) and practice (improved performance versus common synchronization tools like rsync).
Reliable data storage in next-gen memories: New memories have revolutionized the world of storage with their speed and power efficiency. However, modern memories suffer from specific physical limitations that lead to errors and corruption. Novel reliability and error-correcting techniques are critical to the future of these devices. My work involves developing new data representations (balanced sequences or permutations instead of binary strings) and new coding techniques for memories like flash. I am also interested in theoretical frameworks to evaluate broad ranges of error-correction techniques.
I am also very interested in scientific writing and communication. I strongly believe in the importance of clearly and effectively communicating research ideas to a broad and popular audience. Together with my advisor, I have written a book on channel coding for non-volatile memories, a book chapter on advanced error-correction techniques for 3D flash memories, and an expository article on dealing with flash deficiencies.