Vishesh Karwa

Profile Picture of Vishesh Karwa

Vishesh Karwa

  • Fox School of Business and Management

    • Statistics, Operations, and Data Science

      • Assistant Professor

Biography

Dr. Vishesh Karwa is an Assistant Professor of Statistical Science.

He joins the Fox School from The Ohio State University, where he served in the Department of Statistics. He previously served two years as a post-doctoral fellow at Harvard University’s Center for Research and Computation for Society and one year at Carnegie Mellon University.

His research interests include data privacy; causal inference under network interference; social network models; algebraic statistics and computational methods for intractable likelihoods, among others.

Karwa earned his PhD in Statistics from The Pennsylvania State University, where he also attained a Master of Science in Transportation Engineering. He received a Bachelor of Technology in Civil and Environmental Engineering from the Indian Institute of Technology.

Research Interests

  • Statistical foundations of data privacy including differential privacy
  • Causal inference under network interference
  • Experimentation on social media platforms
  • Exponential random graph models for networks and network privacy
  • Computational methods for intractable likelihoods and massive datasets
  • Application of algebraic statistics to log-linear and network models
  • Application of NLP to semantic search and recommendation systems

Courses Taught

Number

Name

Level

STAT 2501

Quantitative Foundations for Data Science

Undergraduate

STAT 5603

Statistical Learning and Data Mining

Graduate

STAT 5604

Experiments: Knowledge by Design

Graduate

Selected Publications

Recent

  • Karwa, V., Gross, E., & Petrovic, S. (2022). Algebraic statistics, tables, and networks: The Fienberg advantage. In Statistics in the Public Interest In Memory of Stephen E. Fienberg. Springer Nature.

  • Karwa, V., Petrović, S., & Bajić, D. (2022). DERGMs: Degeneracy-restricted exponential family random graph models. Network Science, 10(1), 82-110. Cambridge University Press (CUP). doi: 10.1017/nws.2022.5.