Exact and approximate sampling by systematic stochastic search
Author(s)
Tenenbaum, Joshua B.; Jonas, Eric M.; Roy, Daniel; Mansinghka, Vikash K.
DownloadTenenbaum_Exact and.pdf (1.020Mb)
OPEN_ACCESS_POLICY
Open Access Policy
Creative Commons Attribution-Noncommercial-Share Alike
Terms of use
Metadata
Show full item recordAbstract
We introduce _adaptive sequential rejection sampling_, an algorithm for generating exact samples from high-dimensional, discrete distributions, building on ideas from classical AI search. Just as systematic search algorithms like A* recursively build complete solutions from partial solutions, sequential rejection sampling recursively builds exact samples over high-dimensional spaces from exact samples over lower-dimensional subspaces. Our algorithm recovers widely-used particle filters as an approximate variant without adaptation, and a randomized version of the directed arc consistency algorithm with backtracking when applied to deterministic problems. In this paper, we present the mathematical and algorithmic underpinnings of our approach and measure its behavior on ferromagnetic Isings and other probabilistic graphical models, obtaining exact and approximate samples in a range of situations.
Description
URL to conference proceedings page
Date issued
2009-04Department
Massachusetts Institute of Technology. Department of Brain and Cognitive SciencesJournal
Proceedings of the Twelfth International Conference on Artificial Intelligence and Statistics, JMLR Workshop and Conference Proceedings Volume 5: AISTATS 2009
Publisher
Journal of Machine Learning Research
Citation
Mansinghka, Vikash, et al. "Exact and Approximate Sampling by Systematic Stochastic Search." Journal of Machine Learning Research, Workshop & Conference Proceedings, 5:400-407, 2009.
Version: Author's final manuscript
ISSN
1533-792