Prof. David Gregg is a Professor in Computer Science and Fellow of Trinity College Dublin, the sole college of the University of Dublin. He leads a research group within the university working on software performance, particularly for parallel computers.
His current research focus is the challenge of adapting machine-learning and deep learning techniques to embedded systems, large and small. This research includes parallel algorithm design, domain-specific program generation, efficiently supporting custom precision numeric data, and pruning techniques for deep neural networks. The research addresses a variety of parallel computers including vector, embedded multicore, GPU and FPGAs.
Prof. Gregg has published almost 100 papers in venues such as PLDI, TOPLAS, TSE and TOCS. He has served on the technical program committee of leading conferences such as PACT, ISC, ASPLOS and PLDI. He has supervised 15 PhD students to graduation, and has a long record of collaboration with industry.