Supplementary Material: Ordered Stick-Breaking Prior for Sequential MCMC Inference of Bayesian Nonparametric Models
نویسندگان
چکیده
We discuss some relevant mathematical background first in Section S.2, those are directly used in the paper or proofs. We include some examples and properties related to OSBP and PPFs of OSBP in Section S.3. There are two lemmas and four theorems in the paper. We prove them here in Sections S.4, S.5, S.6 related to OSBP, PPF ofOSBP and SUMO respectively. We additionally provide one theorem (Theorem A) and a lemma (Lemma 1) which are strongly related to OSBP but could not be included in the paper due to space constraint. Then we provide construction of dependency over mini-batches using OSBP on PYP, SBP, HDP in Section S.7. Finally we give inference details for DPMM for text datasets in Section S.8.
منابع مشابه
Ordered Stick-Breaking Prior for Sequential MCMC Inference of Bayesian Nonparametric Models
This paper introduces ordered stick-breaking process (OSBP), where the atoms in a stickbreaking process (SBP) appear in order. The choice of weights on the atoms of OSBP ensure that; (1) probability of adding new atoms exponentially decrease, and (2) OSBP, though non-exchangeable, admit predictive probability functions (PPFs). In a Bayesian nonparametric (BNP) setting, OSBP serves as a natural ...
متن کاملGamma Processes, Stick-Breaking, and Variational Inference
While most Bayesian nonparametric models in machine learning have focused on the Dirichlet process, the beta process, or their variants, the gamma process has recently emerged as a useful nonparametric prior in its own right. Current inference schemes for models involving the gamma process are restricted to MCMC-based methods, which limits their scalability. In this paper, we present a variatio...
متن کاملVariational Inference for Sequential Distance Dependent Chinese Restaurant Process
Recently proposed distance dependent Chinese Restaurant Process (ddCRP) generalizes extensively used Chinese Restaurant Process (CRP) by accounting for dependencies between data points. Its posterior is intractable and so far only MCMC methods were used for inference. Because of very different nature of ddCRP no prior developments in variational methods for Bayesian nonparametrics are appliable...
متن کاملBayesian nonparametric regression with varying residual density.
We consider the problem of robust Bayesian inference on the mean regression function allowing the residual density to change flexibly with predictors. The proposed class of models is based on a Gaussian process prior for the mean regression function and mixtures of Gaussians for the collection of residual densities indexed by predictors. Initially considering the homoscedastic case, we propose ...
متن کاملBayesian nonparametric mixed random utility models
Wepropose amixedmultinomial logit model, with themixing distribution assigned a general (nonparametric) stick-breaking prior.Wepresent aMarkov chainMonte Carlo (MCMC) algorithm to sample and estimate the posterior distribution of the model’s parameters. The algorithm relies on a Gibbs (slice) sampler that is useful for Bayesian nonparametric (infinite-dimensional) models. The model and algorith...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015