Disentangled Sticky Hierarchical Dirichlet Process Hidden Markov Model
This is an incremental improvement for researchers in Bayesian nonparametric models and time-series analysis, addressing a specific expressiveness issue in existing methods.
The paper tackled the limitation of the sticky HDP-HMM by proposing a disentangled version that separates self-persistence and transition priors, resulting in improved performance on synthetic and real data, including neural and behavioral video analysis.
The Hierarchical Dirichlet Process Hidden Markov Model (HDP-HMM) has been used widely as a natural Bayesian nonparametric extension of the classical Hidden Markov Model for learning from sequential and time-series data. A sticky extension of the HDP-HMM has been proposed to strengthen the self-persistence probability in the HDP-HMM. However, the sticky HDP-HMM entangles the strength of the self-persistence prior and transition prior together, limiting its expressiveness. Here, we propose a more general model: the disentangled sticky HDP-HMM (DS-HDP-HMM). We develop novel Gibbs sampling algorithms for efficient inference in this model. We show that the disentangled sticky HDP-HMM outperforms the sticky HDP-HMM and HDP-HMM on both synthetic and real data, and apply the new approach to analyze neural data and segment behavioral video data.