Saved in:
Bibliographic Details
Main Authors: Busa-Fekete, Robert, Syed, Umar
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.13804
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912334446854144
author Busa-Fekete, Robert
Syed, Umar
author_facet Busa-Fekete, Robert
Syed, Umar
contents We present new algorithms for estimating and testing \emph{collision probability}, a fundamental measure of the spread of a discrete distribution that is widely used in many scientific fields. We describe an algorithm that satisfies $(α, β)$-local differential privacy and estimates collision probability with error at most $ε$ using $\tilde{O}\left(\frac{\log(1/β)}{α^2 ε^2}\right)$ samples for $α\le 1$, which improves over previous work by a factor of $\frac{1}{α^2}$. We also present a sequential testing algorithm for collision probability, which can distinguish between collision probability values that are separated by $ε$ using $\tilde{O}(\frac{1}{ε^2})$ samples, even when $ε$ is unknown. Our algorithms have nearly the optimal sample complexity, and in experiments we show that they require significantly fewer samples than previous methods.
format Preprint
id arxiv_https___arxiv_org_abs_2504_13804
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Near-optimal algorithms for private estimation and sequential testing of collision probability
Busa-Fekete, Robert
Syed, Umar
Machine Learning
Artificial Intelligence
We present new algorithms for estimating and testing \emph{collision probability}, a fundamental measure of the spread of a discrete distribution that is widely used in many scientific fields. We describe an algorithm that satisfies $(α, β)$-local differential privacy and estimates collision probability with error at most $ε$ using $\tilde{O}\left(\frac{\log(1/β)}{α^2 ε^2}\right)$ samples for $α\le 1$, which improves over previous work by a factor of $\frac{1}{α^2}$. We also present a sequential testing algorithm for collision probability, which can distinguish between collision probability values that are separated by $ε$ using $\tilde{O}(\frac{1}{ε^2})$ samples, even when $ε$ is unknown. Our algorithms have nearly the optimal sample complexity, and in experiments we show that they require significantly fewer samples than previous methods.
title Near-optimal algorithms for private estimation and sequential testing of collision probability
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2504.13804