Wednesday Oct 21, 12pm-1pm, [Meeting Link]
Distributed-Memory Parallel Symmetric Nonnegative Matrix Factorization
Advisors: Prof. Richard Vuduc and Prof. Haesun Park
We develop the first distributed-memory parallel implementation of Symmetric Nonnegative Matrix Factorization (SymNMF), a key data analytics kernel for clustering and dimensionality reduction. Our implementation includes two different algorithms for SymNMF, which give comparable results in terms of time and accuracy. The first algorithm is a parallelization of an existing sequential approach that uses solvers for nonsymmetric NMF. The second algorithm is a novel approach based on the Gauss-Newton method. It exploits second-order information without incurring large computational and memory costs. We evaluate the scalability of our algorithms on the Summit system at Oak Ridge National Laboratory, scaling up to 128 nodes (4096 cores) with 70% efficiency. Additionally, we demonstrate our software on an image segmentation task.
Srinivas Eswar is a fifth year PhD student at the Georgia Tech advised by Rich Vuduc and Haesun Park. He received his BE in Computer Science and MSc in Mathematics from Birla Institute of Technology and Science, Pilani. His research interests are in high performance computing, data mining, and constrained matrix and tensor approximations.