Streaming Variational Inference for Bayesian Nonparametric Mixture Models
Alex Tank, Nick Foti, Emily B. Fox

Citation
Alex Tank, Nick Foti, Emily B. Fox. "Streaming Variational Inference for Bayesian Nonparametric Mixture Models". International Conference on Artificial Intelligence and Statistics (AISTATS), 9, May, 2015.

Abstract
In theory, Bayesian nonparametric (BNP) models are well suited to streaming data scenarios due to their ability to adapt model complexity with the observed data. Unfortunately, such benefits have not been fully realized in practice; existing inference algorithms are either not applicable to streaming applications or not extensible to BNP models. For the special case of Dirichlet processes, streaming inference has been considered. However, there is growing interest in more flexible BNP models building on the class of normalized random measures (NRMs). We work within this general framework and present a streaming variational inference algorithm for NRM mixture models. Our algorithm is based on assumed density filtering (ADF), leading straightforwardly to expectation propagation (EP) for large-scale batch inference as well. We demonstrate the ecacy of the algorithm on clustering documents in large, streaming text corpora.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  Alex Tank, Nick Foti, Emily B. Fox. <a
  href="http://www.terraswarm.org/pubs/494.html"
  >Streaming Variational Inference for Bayesian
  Nonparametric Mixture Models</a>,  International
  Conference on Artificial Intelligence and Statistics
  (AISTATS), 9, May, 2015.

• Plain text

  Alex Tank, Nick Foti, Emily B. Fox. "Streaming
  Variational Inference for Bayesian Nonparametric Mixture
  Models".  International Conference on Artificial
  Intelligence and Statistics (AISTATS), 9, May, 2015.

• BibTeX

  @inproceedings{TankFotiFox15_StreamingVariationalInferenceForBayesianNonparametric,
      author = {Alex Tank and Nick Foti and Emily B. Fox},
      title = {Streaming Variational Inference for Bayesian
                Nonparametric Mixture Models},
      booktitle = { International Conference on Artificial
                Intelligence and Statistics (AISTATS)},
      day = {9},
      month = {May},
      year = {2015},
      abstract = {In theory, Bayesian nonparametric (BNP) models are
                well suited to streaming data scenarios due to
                their ability to adapt model complexity with the
                observed data. Unfortunately, such benefits have
                not been fully realized in practice; existing
                inference algorithms are either not applicable to
                streaming applications or not extensible to BNP
                models. For the special case of Dirichlet
                processes, streaming inference has been
                considered. However, there is growing interest in
                more flexible BNP models building on the class of
                normalized random measures (NRMs). We work within
                this general framework and present a streaming
                variational inference algorithm for NRM mixture
                models. Our algorithm is based on assumed density
                filtering (ADF), leading straightforwardly to
                expectation propagation (EP) for large-scale batch
                inference as well. We demonstrate the ecacy of
                the algorithm on clustering documents in large,
                streaming text corpora.},
      URL = {http://terraswarm.org/pubs/494.html}
  }
  

Posted by Barb Hoversten on 9 Feb 2015.
Groups: services