Ritwik’s Academic Webpage
Hi, I’m Ritwik Vashistha, a final-year PhD candidate in Statistics at the University of Texas at Austin. I develop statistical machine-learning methods for reliable inference under imperfect conditions, in particular when data is noisy, biased or when models may be miscalibrated. My work lies at the intersection of statistics, trustworthy machine learning, and scientific applications.
I enjoy developing statistically rigorous methods motivated by scientific and other real-world applications and turning these methods into practical, reproducible, and accessible software for researchers and practitioners. You can explore my research, software, and publications.
At UT, I am a member of the CosmicAI Institute, where I am advised by Dr. Arya Farahi.
News
- June 2026 Two new preprints on simulation-based inference: OASIS on observation-aware inference with selection biases, and Nonparametric Deconvolution on density deconvolution and denoising.
- May 2026 Won International Society of Bayesian Analysis Travel Award for presenting research at ISBA 2026
- March 2026 Awarded Outstanding Graduate Research Fellowship, covering funding for academic year 2026-2027.
- Feb 2026 New paper on measurement error out on arXiv.
- May 2025 Presented work on trustworthy ML at AISTATS 2025.
- May 2025 Won Keller Award for demonstrating exceptional leadership skills and supporting others during PhD.
- Aug 2024 Started third year as a member of the CosmicAI Institute.
Selected Publications
Convolutional Maximum Mean Discrepancy for Inference in Noisy Data
arXiv preprint, 2026
I-trustworthy Models: A Framework for Trustworthiness Evaluation of Probabilistic Classifiers
AISTATS (PMLR), 2025
U-trustworthy Models: Reliability, Competence, and Confidence in Decision-Making
AAAI Conference on Artificial Intelligence, 2024
Outside Research
Outside research, I use my personal website as a creative outlet, where I share my experiences and photographs from my travels.