DiFacto
Authors:
Open Access
- 8 February 2016
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 377-386
Abstract
Factorization Machines offer good performance and useful embeddings of data. However, they are costly to scale to large amounts of data and large numbers of features. In this paper we describe DiFacto, which uses a refined Factorization Machine model with sparse memory adaptive constraints and frequency adaptive regularization. We show how to distribute DiFacto over multiple machines using the Parameter Server framework by computing distributed subgradients on minibatches asynchronously. We analyze its convergence and demonstrate its efficiency in computational advertising datasets with billions examples and features.Keywords
Funding Information
- Adobe (Recommender Systems)
- National Science Foundation (Distributed Optimization)
This publication has 12 references indexed in Scilit:
- Scaling Distributed Machine Learning with the Parameter Server2014
- Ad click prediction2013
- Distributed large-scale natural graph factorization2013
- Stochastic First- and Zeroth-Order Methods for Nonconvex Stochastic ProgrammingSIAM Journal on Optimization, 2013
- An architecture for parallel topic modelsProceedings of the VLDB Endowment, 2010
- Pairwise interaction tensor factorization for personalized tag recommendation2010
- IntervalRank2010
- Time-Variant Factorization Models2010
- Lessons from the Netflix prize challengeACM SIGKDD Explorations Newsletter, 2007
- Regression Shrinkage and Selection Via the LassoJournal of the Royal Statistical Society Series B: Statistical Methodology, 1996