Protected test-time adaptation via online entropy matching: A betting approach
We present a novel approach for test-time adaptation via online self-training, consisting of
two components. First, we introduce a statistical framework that detects distribution shifts in …
two components. First, we introduce a statistical framework that detects distribution shifts in …
[PDF][PDF] Online Multivariate Changepoint Detection: Leveraging Links With Computational Geometry
L Pishchagina, G Romano, P Fearnhead… - arXiv preprint arXiv …, 2023 - researchgate.net
The increasing volume of data streams poses significant computational challenges for
detecting changepoints online. Likelihood-based methods are effective, but their …
detecting changepoints online. Likelihood-based methods are effective, but their …
Adaptive Learn-then-Test: Statistically Valid and Efficient Hyperparameter Selection
We introduce adaptive learn-then-test (aLTT), an efficient hyperparameter selection
procedure that provides finite-sample statistical guarantees on the population risk of AI …
procedure that provides finite-sample statistical guarantees on the population risk of AI …
Multiple testing in multi-stream sequential change detection
S Dandapanthula, A Ramdas - arXiv preprint arXiv:2501.04130, 2025 - arxiv.org
Multi-stream sequential change detection involves simultaneously monitoring many streams
of data and trying to detect when their distributions change, if at all. Here, we theoretically …
of data and trying to detect when their distributions change, if at all. Here, we theoretically …
Design and Detection of Controller Manipulation Attack on RIS Assisted Communication
SS Acharjee, A Chattopadhyay - IEEE Journal on Selected …, 2024 - ieeexplore.ieee.org
In recent years, research on signal and information theory from the electromagnetics
viewpoint has drawn significant attention, mostly due to its potential use in various …
viewpoint has drawn significant attention, mostly due to its potential use in various …