Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Feng, Qing, Daulton, Samuel, Letham, Benjamin, Balandat, Maximilian, Bakshy, Eytan
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2506.18744
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911029339881472
author Feng, Qing
Daulton, Samuel
Letham, Benjamin
Balandat, Maximilian
Bakshy, Eytan
author_facet Feng, Qing
Daulton, Samuel
Letham, Benjamin
Balandat, Maximilian
Bakshy, Eytan
contents Online experiments in internet systems, also known as A/B tests, are used for a wide range of system tuning problems, such as optimizing recommender system ranking policies and learning adaptive streaming controllers. Decision-makers generally wish to optimize for long-term treatment effects of the system changes, which often requires running experiments for a long time as short-term measurements can be misleading due to non-stationarity in treatment effects over time. The sequential experimentation strategies--which typically involve several iterations--can be prohibitively long in such cases. We describe a novel approach that combines fast experiments (e.g., biased experiments run only for a few hours or days) and/or offline proxies (e.g., off-policy evaluation) with long-running, slow experiments to perform sequential, Bayesian optimization over large action spaces in a short amount of time.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18744
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Experimenting, Fast and Slow: Bayesian Optimization of Long-term Outcomes with Online Experiments
Feng, Qing
Daulton, Samuel
Letham, Benjamin
Balandat, Maximilian
Bakshy, Eytan
Machine Learning
Online experiments in internet systems, also known as A/B tests, are used for a wide range of system tuning problems, such as optimizing recommender system ranking policies and learning adaptive streaming controllers. Decision-makers generally wish to optimize for long-term treatment effects of the system changes, which often requires running experiments for a long time as short-term measurements can be misleading due to non-stationarity in treatment effects over time. The sequential experimentation strategies--which typically involve several iterations--can be prohibitively long in such cases. We describe a novel approach that combines fast experiments (e.g., biased experiments run only for a few hours or days) and/or offline proxies (e.g., off-policy evaluation) with long-running, slow experiments to perform sequential, Bayesian optimization over large action spaces in a short amount of time.
title Experimenting, Fast and Slow: Bayesian Optimization of Long-term Outcomes with Online Experiments
topic Machine Learning
url https://arxiv.org/abs/2506.18744