2014 AReliableEffectiveTerascaleLine

• (Agarwal et al., 2014) ⇒ Alekh Agarwal, Olivier Chapelle, Miroslav Dudík, and John Langford. (2014). “A Reliable Effective Terascale Linear Learning System.” In: The Journal of Machine Learning Research, 15(1).

Subject Headings: Distributed Machine Learning System; Vowpal Wabbit Open Source Project.

Cited By

• http://scholar.google.com/scholar?q=%222014%22+A+Reliable+Effective+Terascale+Linear+Learning+System
• http://dl.acm.org/citation.cfm?id=2627435.2638571&preflayout=flat#citedby

Quotes

Author Keywords

• Distributed machine learning; Hadoop; AllReduce; repeated online averaging; distributed L-BFGS

Abstract

We present a system and a set of techniques for learning linear predictors with convex losses on terascale data sets, with trillions of features, billions of training examples and millions of parameters in an hour using a cluster of 1000 machines. Individually none of the component techniques are new, but the careful synthesis required to obtain an efficient implementation is. The result is, up to our knowledge, the most scalable and efficient linear learning system reported in the literature. We describe and thoroughly evaluate the components of the system, showing the importance of the various design choices.

References

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2014 AReliableEffectiveTerascaleLine	Olivier Chapelle John Langford Alekh Agarwal Miroslav Dudík			A Reliable Effective Terascale Linear Learning System