Coresets for scalable Bayesian logistic regression
Name
NIPS-2016-coresets-for-scalable-bayesian-logistic-regression-Paper.pdf
Description
Published version
Size
1.45 MB
Format
Adobe PDF
Checksum (MD5)
10bf2051f266be79cd5ad4242c6c76fa
Author(s)
Huggins, Jonathan H.
Campbell, Trevor David
Broderick, Tamara A
Date Issued
December 2016
Journal
30th Conference on Neural Information Processing Systems (NIPS 2016)
Publisher
Curran
Citation
Huggins, Jonathan H. et al. “Coresets for scalable Bayesian logistic regression.” Paper presented at the 30th Conference on Neural Information Processing Systems (NIPS 2016), Bacelona, Spain, December 5-10 2016, Curran © 2016 The Author(s)
Version
Final published version
Abstract
The use of Bayesian methods in large-scale data settings is attractive because of the rich hierarchical models, uncertainty quantification, and prior specification they provide. Standard Bayesian inference algorithms are computationally expensive, however, making their direct application to large datasets difficult or infeasible. Recent work on scaling Bayesian inference has focused on modifying the underlying algorithms to, for example, use only a random data subsample at each iteration. We leverage the insight that data is often redundant to instead obtain a weighted subset of the data (called a coreset) that is much smaller than the original dataset. We can then use this small coreset in any number of existing posterior inference algorithms without modification. In this paper, we develop an efficient coreset construction algorithm for Bayesian logistic regression models. We provide theoretical guarantees on the size and approximation quality of the coreset - both for fixed, known datasets, and in expectation for a wide class of data generative models. Crucially, the proposed approach also permits efficient construction of the coreset in both streaming and parallel settings, with minimal additional effort. We demonstrate the efficacy of our approach on a number of synthetic and real-world datasets, and find that, in practice, the size of the coreset is independent of the original dataset size. Furthermore, constructing the coreset takes a negligible amount of time compared to that required to run MCMC on it.
MIT Department
Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://papers.nips.cc/paper/2016/hash/2b0f658cbffd284984fb11d90254081f-Abstract.html
