Sampling RTB transactions in an online machine learning setting

We (the machine learning team at Jampp) strive to predict click-through rates (CTR) and conversion rates (CVR) for the real-time bidding (RTB) online advertising market by means of an in-house online machine learning platform based on a state-of-the-art stochastic gradient descent estimator. Our est...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Pita, Carlos
Formato: Objeto de conferencia
Lenguaje:Inglés
Publicado: 2016
Materias:
Acceso en línea:http://sedici.unlp.edu.ar/handle/10915/56845
http://45jaiio.sadio.org.ar/sites/default/files/AGRANDA-11.pdf
Aporte de:
id I19-R120-10915-56845
record_format dspace
institution Universidad Nacional de La Plata
institution_str I-19
repository_str R-120
collection SEDICI (UNLP)
language Inglés
topic Ciencias Informáticas
Streaming events
demand-side platform
events
spellingShingle Ciencias Informáticas
Streaming events
demand-side platform
events
Pita, Carlos
Sampling RTB transactions in an online machine learning setting
topic_facet Ciencias Informáticas
Streaming events
demand-side platform
events
description We (the machine learning team at Jampp) strive to predict click-through rates (CTR) and conversion rates (CVR) for the real-time bidding (RTB) online advertising market by means of an in-house online machine learning platform based on a state-of-the-art stochastic gradient descent estimator. Our estimation framework has already been covered in a previous paper, so here we want to focus on some peripheral aspects of our platform that, in spite of being of a somewhat ancillary nature, nevertheless tend to dominate development efforts and overall system complexity; namely, in order to feed the learning system we first need to sample a very high-volume stream of out-of-order and scattered-in-time events and consolidate them into a sequence of observations representing the underlying market transactions, each observation composed of a set of features and a response, from which the estimator is ultimately able to learn. This paper is written in a down-to-earth fashion: we describe a number of particular difficulties the general problem of sampling in an online high-volume setting poses and then we present our concrete answers to those difficulties based on real, hands-on, experience.
format Objeto de conferencia
Objeto de conferencia
author Pita, Carlos
author_facet Pita, Carlos
author_sort Pita, Carlos
title Sampling RTB transactions in an online machine learning setting
title_short Sampling RTB transactions in an online machine learning setting
title_full Sampling RTB transactions in an online machine learning setting
title_fullStr Sampling RTB transactions in an online machine learning setting
title_full_unstemmed Sampling RTB transactions in an online machine learning setting
title_sort sampling rtb transactions in an online machine learning setting
publishDate 2016
url http://sedici.unlp.edu.ar/handle/10915/56845
http://45jaiio.sadio.org.ar/sites/default/files/AGRANDA-11.pdf
work_keys_str_mv AT pitacarlos samplingrtbtransactionsinanonlinemachinelearningsetting
bdutipo_str Repositorios
_version_ 1764820477685006336