Indexado en

Enlace externo

Licencia y uso

Citaciones

Altmetrics

Análisis de autorías institucional

Cuesta-Infante AAutor o Coautor

Compartir

Sample, estimate, tune: Scaling Bayesian auto-tuning of data science pipelines

Publicado en:Proceedings - 2017 International Conference On Data Science And Advanced Analytics, Dsaa 2017. 2018-January 361-372 - 2017-01-01 2018-January(), DOI: 10.1109/DSAA.2017.82

Autores: Anderson, Alec; Dubois, Sebastien; Cuesta-Infante, Alfredo; Veeramachaneni, Kalyan

Afiliaciones

Resumen

In this paper, we describe a system for sequential hyperparameter optimization that scales to work with complex pipelines and large datasets. Currently, the state-of-the-art in hyperparameter optimization improves on randomized and grid search by using sequential Bayesian optimization to explore the space of hyperparameters in a more informed way. These methods, however, are not scalable, as the entire data science pipeline still must be evaluated on all the data. By designing a sub sampling based approach to estimate pipeline performance, along with a distributed evaluation system, we provide a scalable solution, which we illustrate using complex image and text data pipelines. For three pipelines, we show that we are able to gain similar performance improvements, but by computing on substantially less data. © 2017 IEEE.

Palabras clave