Salvato in:
Dettagli Bibliografici
Autori principali: Yao, Leon, Li, Paul Yiming, Lu, Jiannan
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2401.14549
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914653431398400
author Yao, Leon
Li, Paul Yiming
Lu, Jiannan
author_facet Yao, Leon
Li, Paul Yiming
Lu, Jiannan
contents In accordance with the principle of "data minimization", many internet companies are opting to record less data. However, this is often at odds with A/B testing efficacy. For experiments with units with multiple observations, one popular data minimizing technique is to aggregate data for each unit. However, exact quantile estimation requires the full observation-level data. In this paper, we develop a method for approximate Quantile Treatment Effect (QTE) analysis using histogram aggregation. In addition, we can also achieve formal privacy guarantees using differential privacy.
format Preprint
id arxiv_https___arxiv_org_abs_2401_14549
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Privacy-preserving Quantile Treatment Effect Estimation for Randomized Controlled Trials
Yao, Leon
Li, Paul Yiming
Lu, Jiannan
Methodology
In accordance with the principle of "data minimization", many internet companies are opting to record less data. However, this is often at odds with A/B testing efficacy. For experiments with units with multiple observations, one popular data minimizing technique is to aggregate data for each unit. However, exact quantile estimation requires the full observation-level data. In this paper, we develop a method for approximate Quantile Treatment Effect (QTE) analysis using histogram aggregation. In addition, we can also achieve formal privacy guarantees using differential privacy.
title Privacy-preserving Quantile Treatment Effect Estimation for Randomized Controlled Trials
topic Methodology
url https://arxiv.org/abs/2401.14549