University of Sussex
Browse
icml2014c2_bratieres14.pdf (847.58 kB)

Scalable Gaussian process structured prediction for grid factor graph applications

Download (847.58 kB)
conference contribution
posted on 2023-06-08, 21:11 authored by Sebastien Bratieres, Novi QuadriantoNovi Quadrianto, Sebastian Nowozin, Zoubin Ghahramani
Structured prediction is an important and well studied problem with many applications across machine learning. GPstruct is a recently proposed structured prediction model that offers appealing properties such as being kernelised, non-parametric, and supporting Bayesian inference (Bratières et al. 2013). The model places a Gaussian process prior over energy functions which describe relationships between input variables and structured output variables. However, the memory demand of GPstruct is quadratic in the number of latent variables and training runtime scales cubically. This prevents GPstruct from being applied to problems involving grid factor graphs, which are prevalent in computer vision and spatial statistics applications. Here we explore a scalable approach to learning GPstruct models based on ensemble learning, with weak learners (predictors) trained on subsets of the latent variables and bootstrap data, which can easily be distributed. We show experiments with 4M latent variables on image segmentation. Our method outperforms widely-used conditional random field models trained with pseudo-likelihood. Moreover, in image segmentation problems it improves over recent state-of-the-art marginal optimisation methods in terms of predictive performance and uncertainty calibration. Finally, it generalises well on all training set sizes.

History

Publication status

  • Published

File Version

  • Published version

Journal

Proceedings of the 32nd International Conference on Machine Learning; Beijing, China; 21 - 26 June 2014

ISSN

1938-7228

Publisher

JMLR

Issue

2

Volume

32

Page range

334-342

Event name

International Conference on Machine Learning (ICML)

Event location

Beijing

Event type

conference

Department affiliated with

  • Informatics Publications

Notes

Proceedings of the 31 st International Conference on Machine Learning, Beijing, China, 2014

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2015-06-18

First Open Access (FOA) Date

2015-06-18

First Compliant Deposit (FCD) Date

2015-06-18

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC