Memory-efficient inference in dynamic graphical models using multiple cores

[edit]

Galen Andrew, Jeff Bilmes ;
Proceedings of the Fifteenth International Conference on Artificial Intelligence and Statistics, PMLR 22:47-53, 2012.

Abstract

We introduce the archipelagos algorithm for memory-efficient multi-core inference in dynamic graphical models. By making use of several processors running in parallel, the archipelagos algorithm uses exponentially less memory compared to basic forward-backward message passing algorithms (O(log T) compared to O(T) on sequences of length T) and, under often-satisfied assumptions on the relative speed of passing forward and backward messages, runs no slower. We also describe a simple variant of the algorithm that achieves a factor of two speedup over forward-backward on a single core. Experiments with our implementation of archipelagos for the computation of posterior marginal probabilities in an HMM validate the space/time complexity analysis: using four cores, the required memory on our test problem was reduced from 8 GB to 319 KB (a factor of 25000) relative to forward-backward, but completed in essentially the same time. The archipelagos algorithm applies to any dynamic graphical model, including dynamic Bayesian networks, conditional random fields, and hidden conditional random fields.

Related Material