Timezone: »
Poster
Hamming Ball Auxiliary Sampling for Factorial Hidden Markov Models
Michalis Titsias · Christopher Yau
We introduce a novel sampling algorithm for Markov chain Monte Carlo-based Bayesian inference for factorial hidden Markov models. This algorithm is based on an auxiliary variable construction that restricts the model space allowing iterative exploration in polynomial time. The sampling approach overcomes limitations with common conditional Gibbs samplers that use asymmetric updates and become easily trapped in local modes. Instead, our method uses symmetric moves that allows joint updating of the latent sequences and improves mixing. We illustrate the application of the approach with simulated and a real data example.
Author Information
Michalis Titsias (DeepMind)
Christopher Yau (University of Oxford)
More from the Same Authors
-
2017 Workshop: Advances in Approximate Bayesian Inference »
Francisco Ruiz · Stephan Mandt · Cheng Zhang · James McInerney · James McInerney · Dustin Tran · Dustin Tran · David Blei · Max Welling · Tamara Broderick · Michalis Titsias -
2017 Poster: Testing and Learning on Distributions with Symmetric Noise Invariance »
Ho Chung Law · Christopher Yau · Dino Sejdinovic -
2016 Poster: One-vs-Each Approximation to Softmax for Scalable Estimation of Probabilities »
Michalis Titsias -
2016 Poster: The Generalized Reparameterization Gradient »
Francisco Ruiz · Michalis Titsias · David Blei -
2015 Poster: Local Expectation Gradients for Black Box Variational Inference »
Michalis Titsias · Miguel Lázaro-Gredilla -
2014 Spotlight: Hamming Ball Auxiliary Sampling for Factorial Hidden Markov Models »
Michalis Titsias · Christopher Yau -
2013 Poster: Variational Inference for Mahalanobis Distance Metrics in Gaussian Process Regression »
Michalis Titsias · Miguel Lazaro-Gredilla
