Rajarshi Guhaniyogi
Associate Professor, Department of Statistics
Texas A&M University
My research interests lie broadly in the development of Bayesian parametric and non-parametric methodology in complex biomedical and machine learning applications. My ongoing research focuses on Bayesian tensor regression, Bayesian regression with heterogeneous objects, Bayesian data Sketching with random compression matrices, distributed Bayesian inference for massive structured data, Bayesian high dimensional regression, deep learning, federated learning, manifold regression, online Bayesian learning with streaming data, spatial and spatio-temporal modeling for big data.
I strongly believe that the improvement of statistical methods is intrinsically tied to addressing complex real-world problems. Collaboration with scientists from various fields plays a pivotal role in achieving this objective, and I find great fulfillment in working alongside both domain experts and fellow researchers in statistics and methodology. My ongoing methodological research threads are directly motivated by strong collaboration with neuroscientists on multi-modal neuroimaing Data, and with environmental and forestry scientists on remote sensing data.
I am honored to receive the Early Investigator Award in 2023 by the American Statistical Association, Section on Statistics and Environment (ENVR) "For exceptional contributions to statistical methodology for Bayesian inference and machine learning methods through rich hierarchical frameworks for high-dimensional environmental data, for student mentoring and for service to the profession." I am also a recipient of the Early Career Award for Statistics and Data Sciences (ECASDS) in 2023 from the International Indian Statistical Association (IISA). I have also been the recipient of the Hellman Fellowship in 2016 awarded by the University of California, Distinguished Student Paper Award in 2012 by ENAR and JSM Student Paper Competition in 2012 by the American Statistical Association, Section on Statistics and Environment (ENVR).
My research is supported by National Institute of Health (R01 award - score of 1 percentile), National Science Foundation (Division of Mathematical Sciences), Office of Naval Research and other federally funded projects with me as PI / Co-PI.
NEWS:
(February 2024) Raj receives The Early Career Award for Statistics and Data Sciences (ECASDS) for 2023 from the International Indian Statistical Association (IISA).
(October 2023) Invited talk at Los Alamos National Laboratories on robust distributed inference and computation for large simulators.
(September 2023) Raj receives a new NIH R01 (as a Co-PI/MPI) grant to study Bayesian multi-object data modeling. The proposal scored 1 percentile in the ASPA Study Section.
(September 2023) Invited talk at the Technometrics Journal Session at ENBIS-23 (Valencia, Spain) on robust distributed inference and computation.
(August 2023) Article on Bayesian covariate-dependent clustering of undirected networks with brain imaging data is accpted subject to minor revision in Technometrics.
(July 2023) Invited talk at EcoStat (virtual) on Bayesian multi-object regression.
(July 2023) Invited talk at ISI World Statistics Congress (Ottawa, Canada) on Bayesian multi-object regression.
(July 2023) Article on sketching in Bayesian high-dimensional regression with big data using Gaussian scale mixture priors is accpted subject to minor revision in JMLR.
(June 2023) Invited talk at IISA Annual Conference (Golden, CO) on robust distributed inference and computation.
(June 2023) Raj Co-chairs scientific committee of IISA Annual Conference (Golden, CO).
(May 2023) Invited talk at the Technometrics Journal Session at Spring Research Conference (Banff, Canada) on robust distributed inference and computation.
(May 2023) Invited talk at Statistical Methods in Imaging Conference 2023 (Minneapolis, MN) on Bayesian multi-object regression.
(May 2023) Raj receives a grant from the Los Alamos National Laboratories to study robust Bayesian distributed computation in large scale simulator data.
(May 2023) Raj Co-organizes NSF-sponsored CBMS conference in Texas A&M (College station, TX) on the foundation of causal graphical model and structural discovery.
(April 2023) Invited talk at Marquette University (Milwaukee, WI) on Bayesian multi-object regression.
(April 2023) Invited talk at Medical College of Wisconsin (Milwaukee, WI) on Bayesian multi-object regression.
(April 2023) Raj receives the prestigious Early Investigator Award from the American Statistical Association, Section on Statistics and Environment (ENVR).
(March 2023) Article on Bayesian tensor regression using the Tucker decomposition for sparse spatial modeling has been invited revision from Biostatistics.
(December 2022) Invited talk at IISA Annual Conference (Bangalore, India) on Bayesian high dimensional regression with big n and p.
(December 2022) Invited talk at CMStatistics 2022 (virtual) on Bayesian high dimensional regression with big n and p.
(December 2022) Raj joins the editorial board of Journal of Machine Learning Research.
(September 2022) Raj joins the editorial board of Journal of Computational and Graphical Statistics.
(September 2022) Invited talk at UT Austin Department of Statistics and Data Sciences (Austin, TX) on Bayesian tensor regression and distributed computation.
(August 2022) Raj receives a new NSF-DMS grant to study the use of random compression matrices for efficient Bayesian computation.