|
Abstract:
|
This paper proposes a new efficient merge-split sampler for both conjugate and nonconjugate Dirichlet process mixture (DPM) models. The latest generation of samplers for these Bayesian nonparametric models effectively update the model parameters, but can have difficulty updating the clustering of the data. To overcome this, merge-split samplers have been developed, but until now these have been limited to conjugate or conditionally-conjugate DPM models. This paper proposes a new MCMC sampler, called the sequentially-allocated merge-split (SAMS) sampler. The sampler borrows ideas from sequential importance sampling. Splits are proposed by sequentially allocating observations to one of two split components using allocation probabilities that condition on previously allocated data. The proposed sampler is substantially more efficient than existing conjugate and nonconjugate samplers.
|
