Stochastic Variational Inference for the HDP-HMM

We derive a variational inference algorithm for the HDP-HMM based on the two-level stick breaking construction. This construction has previously been applied to the hierarchical Dirichlet processes (HDP) for mixed membership models, allowing for efficient handling of the coupled weight parameters. However, the same algorithm is not directly applicable to HDP-based infinite hidden Markov models (HDP-HMM) because of extra sequential dependencies in the Markov chain. In this paper we provide a solution to this problem by deriving a variational inference algorithm for the HDP-HMM, as well as its stochastic extension, for which all parameter updates are in closed form. We apply our algorithm to sequential text analysis and audio signal analysis, comparing our results with the beamsampled iHMM, the parametric HMM, and other variational inference approximations.

[1]  Michael I. Jordan,et al.  Variational inference for Dirichlet process mixtures , 2006 .

[2]  Michael I. Jordan,et al.  An Introduction to Variational Methods for Graphical Models , 1999, Machine Learning.

[3]  Michael I. Jordan,et al.  Hierarchical Dirichlet Processes , 2006 .

[4]  Erik B. Sudderth,et al.  Truly Nonparametric Online Variational Inference for Hierarchical Dirichlet Processes , 2012, NIPS.

[5]  Jason Xu,et al.  Stochastic variational inference for hidden Markov models , 2014, NIPS.

[6]  Matthew J. Beal.  Variational algorithms for approximate Bayesian inference , 2003 .

[7]  Radford M. Neal.  Slice Sampling , 2003, The Annals of Statistics.

[8]  J. Sethuraman.  A CONSTRUCTIVE DEFINITION OF DIRICHLET PRIORS , 1991 .

[9]  Lawrence Carin,et al.  Hidden Markov Models With Stick-Breaking Priors , 2009, IEEE Transactions on Signal Processing.

[10]  Chong Wang,et al.  Online Variational Inference for the Hierarchical Dirichlet Process , 2011, AISTATS.

[11]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[12]  Dan Klein,et al.  The Infinite PCFG Using Hierarchical Dirichlet Processes , 2007, EMNLP.

[13]  John W. Paisley,et al.  Markov Mixed Membership Models , 2015, ICML.

[14]  Matthew J. Johnson,et al.  Stochastic Variational Inference for Bayesian Time Series Models , 2014, ICML.

[15]  Michael I. Jordan,et al.  An HDP-HMM for systems with state persistence , 2008, ICML '08.

[16]  Chong Wang,et al.  Stochastic variational inference , 2012, J. Mach. Learn. Res..

[17]  Francis R. Bach,et al.  Online Learning for Latent Dirichlet Allocation , 2010, NIPS.

[18]  Yee Whye Teh,et al.  Beam sampling for the infinite hidden Markov model , 2008, ICML '08.