Selected papers
A complete list can be found in my CV.
-
Confidence horizons
with Chase Mathis.
arXiv
-
Combining e-values via demi-supermartingales
with Jiahao Ming, Aaditya Ramdas, Yi Shen, and Ruodu Wang.
arXiv
-
Multi-armed sequential hypothesis testing by betting
with Michael I. Jordan and Ricardo J. Sandoval.
arXiv
-
Post-hoc large-sample statistical inference
with Ben Chugg, Etienne Gauthier, Michael I. Jordan, and Aaditya Ramdas.
arXiv
-
On nonasymptotic confidence intervals for treatment effects in randomized experiments
with Ricardo J. Sandoval, Sivaraman Balakrishnan, Avi Feller, and Michael I. Jordan.
arXiv
-
Instance-log-optimality of portfolio-based e-processes and their sequential hypothesis tests
with Michael I. Jordan and Ricardo J. Sandoval.
arXiv
-
Nonasymptotic and distribution-uniform Komlós-Major-Tusnády approximation
with Martin Larsson and Aaditya Ramdas.
arXiv
-
Distribution-uniform anytime-valid sequential inference and the Robbins-Siegmund distributions
with Edward H. Kennedy and Aaditya Ramdas.
arXiv
-
Concentration inequalities for strong laws and laws of the iterated logarithm
with Johannes Ruf.
arXiv
-
Distribution-uniform strong laws of large numbers
with Martin Larsson and Aaditya Ramdas.
The Annals of Applied Probability (to appear)
arXiv
-
Time-uniform central limit theory and asymptotic confidence sequences
with David Arbour, Edward H. Kennedy, Aaditya Ramdas, and Ritwik Sinha.
The Annals of Statistics, 2024
arXiv |
journal
-
Anytime-valid off-policy inference for contextual bandits
with Nikos Karampatziakis, Paul Mineiro, Aaditya Ramdas, and Lili Wu.
ACM/IMS Journal of Data Science, 2023
arXiv |
journal
-
Estimating means of bounded random variables by betting
with Aaditya Ramdas.
Journal of the Royal Statistical Society, Series B, 2024
arXiv
|
journal
Discussion paper — Link to discussions + rejoinder
-
Nonparametric extensions of randomized response for private confidence sets
with Aaditya Ramdas and Zhiwei Steven Wu.
International Conference on Machine Learning (ICML), 2023
arXiv
|
proceedings
Selected for oral presentation
-
RiLACS: Risk-limiting audits via confidence sequences
with Aaditya Ramdas and Philip B. Stark.
International Conference for Electronic Voting (E-Vote-ID), 2021
arXiv
|
proceedings
Best paper award
-
Confidence sequences for sampling without replacement
with Aaditya Ramdas.
Advances in Neural Information Processing Systems (NeurIPS), 2020
arXiv | proceedings
Spotlight paper