Exploiting locality in high-dimensional Factorial hidden Markov models

Authors: Lorenzo Rimella, Nick Whiteley

JMLR 2022 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We demonstrate the new algorithms on synthetic examples and a London Underground passenger flow problem, where the factor graph is effectively given by the train network.
Researcher Affiliation Academia Lorenzo Rimella EMAIL Department of Mathematics and Statistics Lancaster University Lancaster, LA1 4YF, UK Nick Whiteley EMAIL Institute for Statistical Science School of Mathematics University of Bristol Bristol, BS8 1TW, UK and the Alan Turing Institute, UK
Pseudocode Yes Algorithm 1 Approximate Bayes update
Open Source Code Yes All the codes are implemented in Python 3 and available on Github (Online code).
Open Datasets Yes Transport For London, the operator of the London Underground, has made publicly available tap data, consisting of a 5% sample of all Oyster card journeys in a week during November 2009 (Transport for London, 2018).
Dataset Splits Yes The data are split into training given by Monday, Tuesday and Wednesday and test consisting of Thursday and Friday.
Hardware Specification Yes We used the University of Bristol s Blue Crystal High Performance Computing machine. The experiments were run on either one or two standard compute nodes each with 2 x 2.6GHz 8-core Intel E5-2670 (Sandy Bridge) chips and 4GB of RAM per core.
Software Dependencies No All the codes are implemented in Python 3 and available on Github (Online code). Only a programming language with a version (Python 3) is mentioned, without specific versioned libraries or solvers.
Experiment Setup Yes We took X t0, 1u and simulated three data sets of length T 500 from the model with parameters: ˆµ0pxvq 1, xv 1, @v P V, tˆppxv, zvquxv,zv PX ˆ0.6 0.4 0.2 0.8 , c 1, σ2 1.