Talita Perciano Contact: tperciano@lbl.gov
The Project. This project explores the use of Probabilistic Graphical Models (PGM) such as Markov Random Fields (MRF) along with deep learning models to tackle image analysis problems at scale. Those two frameworks have been widely used individually with success in the area of image processing. The main goal is to significantly advance the capabilities of modern deep learning by proposing alternative data representations, based on PGMs, and enabling learning in a non-Euclidean space. Recent works propose the combination of PGM with Deep Neural Networks allowing: (1) easier and more efficient PGM optimization (2) incorporating learning into the PGM. Research scope will include not only the mathematical modeling behind these approaches but also code development taking into account optimization for large datasets. This is an immediate start position and the project is expected to last until September 2021. This is a unique opportunity to work side by side and in collaboration with researchers in the Computational Research Division at LBL, and also to apply the developed techniques to a wide range of scientific problems (materials science, biology, and others). Finally, the student(s) working in this project will have access to our supercomputing facility.
Qualifications. I’m looking for Berkeley students (graduate/undergraduate) with a background in (one or more) of the following topics: applied math, graph-based algorithms, probabilistic graphical models, image processing, classical machine learning techniques, and deep learning. You should be able to work in Python and/or C/C++, and have a strong interest in learning deep concepts related to both PGM and deep learning. Familiarity with modern machine learning frameworks such as PyTorch, TensorFlow, and Caffe would be a plus. Previous research work using/developing deep learning methods would be a plus.
Pay. This is a paid position for an LBNL GSRA appointment. There will be opportunities to co-author research papers to be published in high impact venues.
Hours. Part-time during the academic year and full-time in summer.
Expectations for RAs. RAs are expected to work on-site at LBNL or remotely (considered on a case-by-case basis) and respond promptly to emails between Monday and Friday.
How to Apply. You can apply through Handshake or send a resume, informal transcript and short email (Subject: LBNL GSRA appointment) describing why you are interested in the position to Talita Perciano at tperciano@lbl.gov. I will review applications and schedule follow-up interviews with an eye towards completing hiring as soon as possible.