Guardado en:
Detalles Bibliográficos
Autores principales: Golovin, Daniel, Bartok, Gabor, Chen, Eric, Donahue, Emily, Huang, Tzu-Kuo, Kokiopoulou, Efi, Qin, Ruoyan, Sarda, Nikhil, Sybrandt, Justin, Tjeng, Vincent
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2304.13033
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911948632752128
author Golovin, Daniel
Bartok, Gabor
Chen, Eric
Donahue, Emily
Huang, Tzu-Kuo
Kokiopoulou, Efi
Qin, Ruoyan
Sarda, Nikhil
Sybrandt, Justin
Tjeng, Vincent
author_facet Golovin, Daniel
Bartok, Gabor
Chen, Eric
Donahue, Emily
Huang, Tzu-Kuo
Kokiopoulou, Efi
Qin, Ruoyan
Sarda, Nikhil
Sybrandt, Justin
Tjeng, Vincent
contents In many software systems, heuristics are used to make decisions - such as cache eviction, task scheduling, and information presentation - that have a significant impact on overall system behavior. While machine learning may outperform these heuristics, replacing existing heuristics in a production system safely and reliably can be prohibitively costly. We present SmartChoices, a novel approach that reduces the cost to deploy production-ready ML solutions for contextual bandits problems. SmartChoices' interface cleanly separates problem formulation from implementation details: engineers describe their use case by defining datatypes for the context, arms, and feedback that are passed to SmartChoices APIs, while SmartChoices manages encoding & logging data and training, evaluating & deploying policies. Our implementation codifies best practices, is efficient enough for use in low-level applications, and provides valuable production features off the shelf via a shared library. Overall, SmartChoices enables non-experts to rapidly deploy production-ready ML solutions by eliminating many sources of technical debt common to ML systems. Engineers have independently used SmartChoices to improve a wide range of software including caches, batch processing workloads, and UI layouts, resulting in better latency, throughput, and click-through rates.
format Preprint
id arxiv_https___arxiv_org_abs_2304_13033
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle SmartChoices: Augmenting Software with Learned Implementations
Golovin, Daniel
Bartok, Gabor
Chen, Eric
Donahue, Emily
Huang, Tzu-Kuo
Kokiopoulou, Efi
Qin, Ruoyan
Sarda, Nikhil
Sybrandt, Justin
Tjeng, Vincent
Software Engineering
Machine Learning
In many software systems, heuristics are used to make decisions - such as cache eviction, task scheduling, and information presentation - that have a significant impact on overall system behavior. While machine learning may outperform these heuristics, replacing existing heuristics in a production system safely and reliably can be prohibitively costly. We present SmartChoices, a novel approach that reduces the cost to deploy production-ready ML solutions for contextual bandits problems. SmartChoices' interface cleanly separates problem formulation from implementation details: engineers describe their use case by defining datatypes for the context, arms, and feedback that are passed to SmartChoices APIs, while SmartChoices manages encoding & logging data and training, evaluating & deploying policies. Our implementation codifies best practices, is efficient enough for use in low-level applications, and provides valuable production features off the shelf via a shared library. Overall, SmartChoices enables non-experts to rapidly deploy production-ready ML solutions by eliminating many sources of technical debt common to ML systems. Engineers have independently used SmartChoices to improve a wide range of software including caches, batch processing workloads, and UI layouts, resulting in better latency, throughput, and click-through rates.
title SmartChoices: Augmenting Software with Learned Implementations
topic Software Engineering
Machine Learning
url https://arxiv.org/abs/2304.13033