Event

Scalable Low Rank Approximation and Graph Algorithms

Dr. Ramki Kannan
Dr. Ramki Kannan

Abstract: Given an input matrix A, Non-negative Matrix Factorization (NMF) is about determining two non-negative matrices called factors, corresponding to the rows and columns of A respectively,  such that the product of them closely approximates A. This is called “low rank approximation” as the rank of the factors are generally much smaller than the input matrix A.  Also, it is known as dimensionality reduction in Machine Learning Community and inverse problems in scientific community. In recent times, there is a huge interest for non-negative factorizations for data with more than two modes, which can be represented via a generalization of matrices known as tensors. There are wide applications of non-negative matrix and tensor factorization (NTF) in the scientific community such as spectral unmixing, compressing scientific data, scientific visualization, inverse problems, feature engineering for deep learning etc., as the factors are scientifically interpretable and an important cornerstone for explainable AI and representation learning. In this talk, we will look at various constraints on factors along with non-negativity such as symmetric, spatial smoothness, sparsity etc., and challenges in the realization of scalable algorithms. In the latter part of the talk, we will focus on some of the recent graph algorithms on Summit. One of our recent work on Semi-ring based APSP has become the  2020 Gordon bell finalist.
 
Bio: Ramki is the team lead for Computational Artificial Intelligence and Machine Learning(CAIML) in Oak Ridge National Laboratory focusing on large scale data mining and machine learning algorithms on HPC systems and modern architectures with applications from scientific domain and many different internet services. He received his Ph.D in Computer Science from College of Computing, Georgia Institute of Technology(GaTech) advised by Prof. Haesun Park. Prior to Ph.D., worked on Data Analytics group at IBM TJ Watson Research Center and was an IBM Master Inventor. Under the advise of Prof. Narahari, in 2008, graduated from Indian Institute of Science with Master's in Engineering. 

Join Microsoft Teams Meeting

+1 865-276-6990   United States, Knoxville (Toll)

Conference ID: 100 522 869#

Last Updated: July 20, 2020 - 1:18 pm