Timezone: »

 
Poster
Scalable Adaptation of State Complexity for Nonparametric Hidden Markov Models
Michael Hughes · William Stephenson · Erik Sudderth

Mon Dec 07 04:00 PM -- 08:59 PM (PST) @ 210 C #29
in Poster Session »

Bayesian nonparametric hidden Markov models are typically learned via fixed truncations of the infinite state space or local Monte Carlo proposals that make small changes to the state space. We develop an inference algorithm for the sticky hierarchical Dirichlet process hidden Markov model that scales to big datasets by processing a few sequences at a time yet allows rapid adaptation of the state space cardinality. Unlike previous point-estimate methods, our novel variational bound penalizes redundant or irrelevant states and thus enables optimization of the state space. Our birth proposals use observed data statistics to create useful new states that escape local optima. Merge and delete proposals remove ineffective states to yield simpler models with more affordable future computations. Experiments on speaker diarization, motion capture, and epigenetic chromatin datasets discover models that are more compact, more interpretable, and better aligned to ground truth segmentations than competitors. We have released an open-source Python implementation which can parallelize local inference steps across sequences.

[ PDF] 

Author Information

Michael Hughes (Brown University)
William Stephenson (Brown University)
Erik Sudderth (Brown University)

More from the Same Authors