Exponential Stochastic Cellular Automata for Massively Parallel Inference

@InProceedings{pmlr-v51-zaheer16,
  title = 	 {Exponential Stochastic Cellular Automata for Massively Parallel Inference},
  author = 	 {Zaheer, Manzil and Wick, Michael and Tristan, Jean-Baptiste and Smola, Alex and Steele, Guy},
  booktitle = 	 {Proceedings of the 19th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {966--975},
  year = 	 {2016},
  editor = 	 {Gretton, Arthur and Robert, Christian C.},
  volume = 	 {51},
  series = 	 {Proceedings of Machine Learning Research},
  address = 	 {Cadiz, Spain},
  month = 	 {09--11 May},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v51/zaheer16.pdf},
  url = 	 {https://proceedings.mlr.press/v51/zaheer16.html},
  abstract = 	 {We propose an embarrassingly parallel, memory efficient inference algorithm for latent variable models in which the complete data likelihood is in the exponential family. The algorithm is a stochastic cellular automaton and converges to a valid maximum a posteriori fixed point. Applied to latent Dirichlet allocation we find that our algorithm is over an order or magnitude faster than the fastest current approaches. A simple C++/MPI implementation on a 20-node Amazon EC2 cluster samples at more than 1 billion tokens per second. We process 3 billion documents and achieve predictive power competitive with collapsed Gibbs sampling and variational inference.}
}