From the 1 of 8 linked papers with an AI index.
8 papers
Non-partitioned e-detectors for nonparametric sequential change detection
Aytijhya Saha, Aaditya Ramdas
The paper proposes non‑partitioned e‑detectors for sequential change detection when both pre‑ and post‑change distributions are unknown, using aggregated e‑processes and showing as…
Causal Inference with Categorical Unobserved Confounder via Mixture Learning
Aytijhya Saha, Stephen Bates, Devavrat Shah
Unobserved confounding is a fundamental challenge for estimating causal effects. To address unobserved confounding, recent literature has turned to two different approaches -- prox…
Distribution-free changepoint localization after sequential change detection
Aytijhya Saha, Aaditya Ramdas
This paper introduces a distribution-free framework for constructing post-detection confidence sets for changepoints after stopping a sequential change detection procedure. It is w…
Post-detection inference for sequential changepoint localization
Aytijhya Saha, Aaditya Ramdas
This paper addresses a fundamental but largely unexplored challenge in sequential changepoint analysis: conducting inference following a detected change. We develop a very general…
Optimal e-variables under constraints
Aytijhya Saha, Aaditya Ramdas
E-variables enable safe and anytime-valid inference, with log-optimal e-variables given by the likelihood ratio of the least favorable distributions (LFDs) when they exist in compo…
Huber-robust likelihood ratio tests for composite nulls and alternatives
Aytijhya Saha, Aaditya Ramdas
We propose an e-value based framework for testing arbitrary composite nulls against composite alternatives, when an fraction of the data can be arbitrarily corrupted. Our test…