tran-hierarchicalsemi-2017.pdf (899.2 kB)
Hierarchical semi-Markov conditional random fields for deep recursive sequential data
journal contribution
posted on 2017-05-01, 00:00 authored by Truyen TranTruyen Tran, Quoc-Dinh Phung, H Bui, Svetha VenkateshSvetha VenkateshWe present the hierarchical semi-Markov conditional random field (HSCRF), a generalisation of linear-chain conditional random fields to model deep nested Markov processes. It is parameterised as a conditional log-linear model and has polynomial time algorithms for learning and inference. We derive algorithms for partially-supervised learning and constrained inference. We develop numerical scaling procedures that handle the overflow problem. We show that when depth is two, the HSCRF can be reduced to the semi-Markov conditional random fields. Finally, we demonstrate the HSCRF on two applications: (i) recognising human activities of daily living (ADLs) from indoor surveillance cameras, and (ii) noun-phrase chunking. The HSCRF is capable of learning rich hierarchical models with reasonable accuracy in both fully and partially observed data cases.
History
Journal
Artificial intelligenceVolume
246Pagination
53 - 85Publisher
ElsevierLocation
Amsterdam, The NetherlandsPublisher DOI
Link to full text
ISSN
0004-3702Language
engPublication classification
C Journal article; C1 Refereed article in a scholarly journalCopyright notice
2017, ElsevierUsage metrics
Categories
Keywords
deep nested sequential processeshierarchical semi-Markov conditionalrandom fieldpartial labellingconstrained inferencenumerical scalingScience & TechnologyTechnologyComputer Science, Artificial IntelligenceComputer ScienceMODELSArtificial Intelligence and Image ProcessingComputation Theory and Mathematics
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC