3.0

credits

Average Course Rating

This course will study recent advances in the systems support for scalable machine learning and artificial intelligence workloads. We will look at recent advances in programming languages, compilers, and operating systems and how they are being implemented in hardware accelerators. Topics will include data sparsity, single-instruction multi-threading, vectorization of memory and processing, and scatter/gather/reduce patterns. The course will alternate between lectures and seminar-style paper discussions. Each student will be expected to conduct and present a substantial research project individually or in teams. Required Course Background: 601.420/620 or equivalent.