论文信息 - Detailed Derivations of Small-Variance Asymptotics for some Hierarchical Bayesian Nonparametric Models

Detailed Derivations of Small-Variance Asymptotics for some Hierarchical Bayesian Nonparametric Models

In this note we provide detailed derivations of two versions of small-variance asymptotics for hierarchical Dirichlet process (HDP) mixture models and the HDP hidden Markov model (HDP-HMM, a.k.a. the infinite HMM). We include derivations for the probabilities of certain CRP and CRF partitions, which are of more general interest.

Matthew J. Johnson | Ardavan Saeedi | Jonathan H. Huggins

[1] Michael I. Jordan,et al. Revisiting k-means: New Algorithms via Bayesian Nonparametrics , 2011, ICML.

[2] Carl E. Rasmussen,et al. Factorial Hidden Markov Models , 1997 .

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

[4] Michael I. Jordan,et al. Small-Variance Asymptotics for Exponential Family Dirichlet Process Mixture Models , 2012, NIPS.

[5] Michael I. Jordan,et al. MAD-Bayes: MAP-based Asymptotic Derivations from Bayes , 2012, ICML.

[6] D. Aldous. Exchangeability and related topics , 1985 .

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

[8] C. Antoniak. Mixtures of Dirichlet Processes with Applications to Bayesian Nonparametric Problems , 1974 .

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