Mathematics of Bigdata Analytics

A comprehensive short course series on the mathematical foundations of big data analytics

Prof. S. Lakshmivarahan Dec. 2025 - Feb. 2026 Hosts: Prof. C. S. Seelamantula, Prof. P. S. Sastry
Prof. S. Lakshmivarahan

Prof. S. Lakshmivarahan

After completing his PhD from the Indian Institute of Science in 1973, S. Lakshmivarahan held faculty and post-doctoral positions at the IIT-Madras, Brown and Yale Universities through 1978. In the Fall of 1978, he joined the School of Computer Science, University of Oklahoma (OU) where he held the position of George Lynn Cross Research Professor since 1995. His research interests are in Applied Mathematics and Computation and includes Data Mining and Analytics, Data Assimilation, Computational Finance, Parallel Computation and Learning Algorithms. He is an author/coauthor of six books in these areas. He was elected as a Fellow of the IEEE in 1993 and a Fellow of ACM in 1995. He has held short-term visiting positions in Japan, China, Taiwan, Thailand, India, Germany, England, Mexico, Brazil, Canada, and USA. Since July 2019, he holds the position of George Lynn Cross Research Professor Emeritus at the School of Computer Science, OU.
Date & Time Tuesdays and Thursdays; 3:30 PM to 5:00 PM
Classroom Room B303, EE Department, IISc
Duration December 2025 - February 2026
Date Title Description
January 13, 2026 Module II Ends

Module II: Concentration of Probability Measures has been completed

January 6, 2026 Module II Begins

Module II: Concentration of Probability Measures begins on January 6, 2026.

January 1, 2026 Module I Completed

Module I: Dimension Reduction has been completed.

Module I: Dimension Reduction

Theory and Applications to Image Processing. Covers Spectral Decomposition, SVD, covariance, geometric projections, and Principal Component Analysis (PCA) with applications in image processing.

6 Lectures

Completed

Module II: Concentration of Probability Measures

Derivation of concentration inequalities (Markov, Chebyshev, Chernoff), sub-Gaussian and sub-exponential variables, Johnson-Lindenstrauss lemma, and Berry-Esseen theorem.

4-5 Lectures

Completed

Module III: Randomized Algorithms for Linear Algebra (RALA)

Sampling-based approaches for matrix operations, matrix sketching, SVD, CUR decomposition, and fast randomized methods for large-scale least squares problems.

4-5 Lectures

Upcoming
Coming Soon

Module IV: Inverse Problems in Empirical Modeling

Solving linear/nonlinear, over/under determined inverse problems using matrix decomposition (LU, Cholesky, QR, SVD) and optimization methods (Gradient, Conjugate gradient, Quasi-Newton).

4-5 Lectures

In Progress