Title :

Trade-offs in Large-Scale Distributed Tuplewise Estimation and Learning

Author(s) :

Vogel, Robin [Auteur]
Laboratoire Traitement et Communication de l'Information [LTCI]
Bellet, Aurelien [Auteur]
Machine Learning in Information Networks [MAGNET]
Clémençon, Stéphan [Auteur]
Laboratoire Traitement et Communication de l'Information [LTCI]
Jelassi, Ons [Auteur]
Laboratoire Traitement et Communication de l'Information [LTCI]
Papa, Guillaume [Auteur]
Laboratoire Traitement et Communication de l'Information [LTCI]

Conference title :

ECML PKDD 2019 - European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases

City :

Würzburg

Country :

Allemagne

Start date of the conference :

2019-09-16

English keyword(s) :

Distributed Machine Learning
Distributed Data Processing
U-Statistics
AUC Optimization

HAL domain(s) :

Informatique [cs]/Apprentissage [cs.LG]
Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

The development of cluster computing frameworks has allowed practitioners to scale out various statistical estimation and machine learning algorithms with minimal programming effort. This is especially true for machine ...
Show more >The development of cluster computing frameworks has allowed practitioners to scale out various statistical estimation and machine learning algorithms with minimal programming effort. This is especially true for machine learning problems whose objective function is nicely separable across individual data points, such as classification and regression. In contrast, statistical learning tasks involving pairs (or more generally tuples) of data points-such as metric learning, clustering or ranking-do not lend themselves as easily to data-parallelism and in-memory computing. In this paper, we investigate how to balance between statistical performance and computational efficiency in such distributed tuplewise statistical problems. We first propose a simple strategy based on occasionally repartitioning data across workers between parallel computation stages, where the number of repartition-ing steps rules the trade-off between accuracy and runtime. We then present some theoretical results highlighting the benefits brought by the proposed method in terms of variance reduction, and extend our results to design distributed stochastic gradient descent algorithms for tuplewise empirical risk minimization. Our results are supported by numerical experiments in pairwise statistical estimation and learning on synthetic and real-world datasets.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL