Sanvesh Srivastava

Assistant Professor of Statistics and Actuarial Science, College of Liberal Arts and Sciences

Contact Information

Primary Office: 219 Schaeffer Hall (SH)
20 East Washington Street
Iowa City, IA 52242
319-335-0824

Office Hours

Monday: 11:30 am - 1:30 pm
Wednesday: 11:30 am - 1:30 pm

Biography

Sanvesh Srivastava is currently an Assistant Professor in the Department of Statistics and Actuarial Science at The University of Iowa. His research aims to develop flexible Bayesian methods and efficient computational algorithms for big data sets, tailored for both their complexity and size. Motivating examples include big data in genomics, medical imaging, and recommender systems. Simultaneously optimizing for the size and complexity is a challenge with current Bayesian methods. He is developing novel and computationally tractable Bayesian methods using principles from machine learning and optimal transportation. Before coming to the University of Iowa, Sanvesh received his Ph.D. in Statistics in August, 2013 from Purdue University, where he also won I.W. Burr Award for "promise of contribution to the profession as evidenced by academic excellence in courses and exams, by the quality of research, and by excellence in teaching and consulting." After Ph.D., he spent two years at Duke University and Statistical and Applied Mathematical Sciences Institute (SAMSI) as a postdoctoral researcher. He has extensive experience in collaborating with scientists and teaching statistics to students from diverse areas and varied expertise.

Education

  • PhD in Statistics, Purdue University, West Lafayette, Indiana, United States, 2013
  • MS in Mathematical Statistics, Purdue University, West Lafayette, Indiana, United States, 2011
  • MSc in Mathematics and Scientific Computing, Indian Institute of Technology Kanpur, Kanpur, India, 2007

Areas of Research Interest

  • Bayesian methods for complex and large data (big data); Computational Statistics; Machine Learning: My current research is motivated by big data in genomics and medical imaging. Bayesian non-parametric methods provide a flexible way to model the complexity of these data that is characterized by nonlinear dependence and low signal strength. This modeling framework also facilitates extensions that are able to capture data-specific patterns. The computational cost of this flexibility scales with the size of the data and the number of parameters used to model the complexity of the data; therefore, Bayesian computations are expensive for big data, which in turn challenges classical Bayesian inferential approaches. As such, my specific research focus is twofold: (a) data analysis with computationally feasible parameter estimation and posterior sampling; and (b) data analysis with only computationally feasible parameter estimation.

Selected Publications

Selected Presentations

Last Modified Date: August 16, 2019