Composing graphical models with neural networks for structured representations and fast inference

Matthew James Johnson; David Duvenaud; Alexander B. Wiltschko; Sandeep R. Datta; Ryan P. Adams

Composing graphical models with neural networks for structured representations and fast inference

Matthew James Johnson, David Duvenaud, Alexander B. Wiltschko, Sandeep R. Datta, Ryan P. Adams

Research output: Contribution to journal › Conference article › peer-review

Abstract

We propose a general modeling and inference framework that combines the complementary strengths of probabilistic graphical models and deep learning methods. Our model family composes latent graphical models with neural network observation likelihoods. For inference, we use recognition networks to produce local evidence potentials, then combine them with the model distribution using efficient message-passing algorithms. All components are trained simultaneously with a single stochastic variational inference objective. We illustrate this framework by automatically segmenting and categorizing mouse behavior from raw depth video, and demonstrate several other example models.

Original language	English (US)
Pages (from-to)	2954-2962
Number of pages	9
Journal	Advances in Neural Information Processing Systems
State	Published - 2016
Externally published	Yes
Event	30th Annual Conference on Neural Information Processing Systems, NIPS 2016 - Barcelona, Spain Duration: Dec 5 2016 → Dec 10 2016

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Information Systems
Signal Processing

Cite this

@article{fb98a64c40ef457992c65fbf257bb20c,

title = "Composing graphical models with neural networks for structured representations and fast inference",

abstract = "We propose a general modeling and inference framework that combines the complementary strengths of probabilistic graphical models and deep learning methods. Our model family composes latent graphical models with neural network observation likelihoods. For inference, we use recognition networks to produce local evidence potentials, then combine them with the model distribution using efficient message-passing algorithms. All components are trained simultaneously with a single stochastic variational inference objective. We illustrate this framework by automatically segmenting and categorizing mouse behavior from raw depth video, and demonstrate several other example models.",

author = "Johnson, {Matthew James} and David Duvenaud and Wiltschko, {Alexander B.} and Datta, {Sandeep R.} and Adams, {Ryan P.}",

note = "Publisher Copyright: {\textcopyright} 2016 NIPS Foundation - All Rights Reserved.; 30th Annual Conference on Neural Information Processing Systems, NIPS 2016 ; Conference date: 05-12-2016 Through 10-12-2016",

year = "2016",

language = "English (US)",

pages = "2954--2962",

journal = "Advances in Neural Information Processing Systems",

issn = "1049-5258",

publisher = "Neural information processing systems foundation",

}

TY - JOUR

T1 - Composing graphical models with neural networks for structured representations and fast inference

AU - Johnson, Matthew James

AU - Duvenaud, David

AU - Wiltschko, Alexander B.

AU - Datta, Sandeep R.

AU - Adams, Ryan P.

PY - 2016

Y1 - 2016

N2 - We propose a general modeling and inference framework that combines the complementary strengths of probabilistic graphical models and deep learning methods. Our model family composes latent graphical models with neural network observation likelihoods. For inference, we use recognition networks to produce local evidence potentials, then combine them with the model distribution using efficient message-passing algorithms. All components are trained simultaneously with a single stochastic variational inference objective. We illustrate this framework by automatically segmenting and categorizing mouse behavior from raw depth video, and demonstrate several other example models.

AB - We propose a general modeling and inference framework that combines the complementary strengths of probabilistic graphical models and deep learning methods. Our model family composes latent graphical models with neural network observation likelihoods. For inference, we use recognition networks to produce local evidence potentials, then combine them with the model distribution using efficient message-passing algorithms. All components are trained simultaneously with a single stochastic variational inference objective. We illustrate this framework by automatically segmenting and categorizing mouse behavior from raw depth video, and demonstrate several other example models.

UR - http://www.scopus.com/inward/record.url?scp=85017433100&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017433100&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85017433100

SN - 1049-5258

SP - 2954

EP - 2962

JO - Advances in Neural Information Processing Systems

JF - Advances in Neural Information Processing Systems

T2 - 30th Annual Conference on Neural Information Processing Systems, NIPS 2016

Y2 - 5 December 2016 through 10 December 2016

ER -

Composing graphical models with neural networks for structured representations and fast inference

Abstract

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this