Efficient Modeling of Latent Information in Supervised Learning using Gaussian Processes

Zhenwen Dai; Mauricio A. Álvarez; Neil D. Lawrence

edit

Back to publications

Efficient Modeling of Latent Information in Supervised Learning using Gaussian Processes

Zhenwen Dai, Mauricio A. Álvarez, Neil D. Lawrence

Advances in Neural Information Processing Systems, Curran Associates, Inc. 30:5131-5139, 2017.

Abstract

Often in machine learning, data are collected as a combination of multiple conditions, e.g., the voice recordings of multiple persons, each labeled with an ID. How could we build a model that captures the latent information related to these conditions and generalize to a new one with few data? We present a new model called Latent Variable Multiple Output Gaussian Processes (LVMOGP) and that allows to jointly model multiple conditions for regression and generalize to a new condition with a few data points at test time. LVMOGP infers the posteriors of Gaussian processes together with a latent space representing the information about different conditions. We derive an efficient variational inference method for LVMOGP, of which the computational complexity is as low as sparse Gaussian processes. We show that LVMOGP significantly outperforms related Gaussian process methods on various tasks with both synthetic and real data.

Links

Cite this Paper

BibTeX


@InProceedings{Dai:supervised17,
  title = 	 {Efficient Modeling of Latent Information in Supervised Learning using {G}aussian Processes},
  author = 	 {Dai, Zhenwen and Álvarez, Mauricio A. and Lawrence, Neil D.},
  booktitle = 	 {Advances in Neural Information Processing Systems},
  pages = 	 {5131--5139},
  year = 	 {2017},
  volume = 	 {30},
  address = 	 {Longbeach, Californa},
  publisher =    {Curran Associates, Inc.},
  pdf = 	 {https://proceedings.neurips.cc/paper/2017/file/1680e9fa7b4dd5d62ece800239bb53bd-Paper.pdf},
  url = 	 {/publications/efficient-modelling-of-latent-information-in-supervised-learning-using-gaussian-processes.html},
  abstract = 	 {Often in machine learning, data are collected as a combination of
multiple conditions, e.g., the voice recordings of multiple persons,
each labeled with an ID. How could we build a model that captures
the latent information related to these conditions and generalize to
a new one with few data? We present a new model called Latent
Variable Multiple Output Gaussian Processes (LVMOGP) and that allows
to jointly model multiple conditions for regression and generalize
to a new condition with a few data points at test time. LVMOGP
infers the posteriors of Gaussian processes together with a latent
space representing the information about different conditions. We
derive an efficient variational inference method for LVMOGP, of
which the computational complexity is as low as sparse Gaussian
processes. We show that LVMOGP significantly outperforms related
Gaussian process methods on various tasks with both synthetic and
real data.
}
}

Endnote

%0 Conference Paper
%T Efficient Modeling of Latent Information in Supervised Learning using Gaussian Processes
%A Zhenwen Dai
%A Mauricio A. Álvarez
%A Neil D. Lawrence
%B Advances in Neural Information Processing Systems
%D 2017	
%F Dai:supervised17
%I Curran Associates, Inc.
%P 5131--5139
%U /publications/efficient-modelling-of-latent-information-in-supervised-learning-using-gaussian-processes.html
%V 30
%X Often in machine learning, data are collected as a combination of
multiple conditions, e.g., the voice recordings of multiple persons,
each labeled with an ID. How could we build a model that captures
the latent information related to these conditions and generalize to
a new one with few data? We present a new model called Latent
Variable Multiple Output Gaussian Processes (LVMOGP) and that allows
to jointly model multiple conditions for regression and generalize
to a new condition with a few data points at test time. LVMOGP
infers the posteriors of Gaussian processes together with a latent
space representing the information about different conditions. We
derive an efficient variational inference method for LVMOGP, of
which the computational complexity is as low as sparse Gaussian
processes. We show that LVMOGP significantly outperforms related
Gaussian process methods on various tasks with both synthetic and
real data.

RIS


TY  - CPAPER
TI  - Efficient Modeling of Latent Information in Supervised Learning using Gaussian Processes
AU  - Zhenwen Dai
AU  - Mauricio A. Álvarez
AU  - Neil D. Lawrence
BT  - Advances in Neural Information Processing Systems
DA  - 2017/12/05	
ID  - Dai:supervised17
PB  - Curran Associates, Inc.
VL  - 30
SP  - 5131
EP  - 5139
L1  - https://proceedings.neurips.cc/paper/2017/file/1680e9fa7b4dd5d62ece800239bb53bd-Paper.pdf
UR  - /publications/efficient-modelling-of-latent-information-in-supervised-learning-using-gaussian-processes.html
AB  - Often in machine learning, data are collected as a combination of
multiple conditions, e.g., the voice recordings of multiple persons,
each labeled with an ID. How could we build a model that captures
the latent information related to these conditions and generalize to
a new one with few data? We present a new model called Latent
Variable Multiple Output Gaussian Processes (LVMOGP) and that allows
to jointly model multiple conditions for regression and generalize
to a new condition with a few data points at test time. LVMOGP
infers the posteriors of Gaussian processes together with a latent
space representing the information about different conditions. We
derive an efficient variational inference method for LVMOGP, of
which the computational complexity is as low as sparse Gaussian
processes. We show that LVMOGP significantly outperforms related
Gaussian process methods on various tasks with both synthetic and
real data.

ER  -

APA


Dai, Z., Álvarez, M.A. & Lawrence, N.D.. (2017). Efficient Modeling of Latent Information in Supervised Learning using Gaussian Processes. Advances in Neural Information Processing Systems 30:5131-5139 Available from /publications/efficient-modelling-of-latent-information-in-supervised-learning-using-gaussian-processes.html.