Hierarchical semi-Markov conditional random fields for deep recursive sequential data

DOI10.1016/J.ARTINT.2017.02.003MaRDI QIDQ2407880zbMATH OpenOpenAlexDBLPWikidataFDO

Authors Truyen Tran, Dinh Phung, Hung Bui, Svetha Venkatesh

Publication date 6 October 2017

Published in Artificial Intelligence (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1009.2009

zbMATH Keywords

constrained inferencedeep nested sequential processeshierarchical semi-Markov conditional random fieldnumerical scalingpartial labelling

Mathematics Subject Classification ID

Applications of Markov chains and discrete-time Markov processes on general state spaces (social mobility, learning theory, industrial processes, etc.) (60J20) Random fields (60G60) Logic in artificial intelligence (68T27)

Abstract: Inspired by the hierarchical hidden Markov models (HHMM), we present the hierarchical semi-Markov conditional random field (HSCRF), a generalisation of embedded undirectedMarkov chains tomodel complex hierarchical, nestedMarkov processes. It is parameterised in a discriminative framework and has polynomial time algorithms for learning and inference. Importantly, we consider partiallysupervised learning and propose algorithms for generalised partially-supervised learning and constrained inference. We demonstrate the HSCRF in two applications: (i) recognising human activities of daily living (ADLs) from indoor surveillance cameras, and (ii) noun-phrase chunking. We show that the HSCRF is capable of learning rich hierarchical models with reasonable accuracy in both fully and partially observed data cases.

Recommendations

• Dynamic conditional random fields: factorized probabilistic models for labeling and segmenting sequence data
• An introduction to conditional random fields
• The hierarchical hidden Markov model: Analysis and applications
• Inference algorithms for pattern-based CRFs on sequence data
• scientific article; zbMATH DE number 6276249

Cites work

• scientific article; zbMATH DE number 6378100 (Why is no real title available?)
• scientific article; zbMATH DE number 1134987 (Why is no real title available?)
• scientific article; zbMATH DE number 1946860 (Why is no real title available?)
• Constructing Free-Energy Approximations and Generalized Belief Propagation Algorithms
• Dynamic conditional random fields: factorized probabilistic models for labeling and segmenting sequence data
• Factor graphs and the sum-product algorithm
• Learning deep architectures for AI
• Reducing the Dimensionality of Data with Neural Networks
• The hierarchical hidden Markov model: Analysis and applications

Cited in

(3)

• scientific article; zbMATH DE number 6378100 (Why is no real title available?)
• Action categorization with modified hidden conditional random field
• Recursive learning for sparse Markov models

Describes a project that uses

Uses Software

• darch

This page was built for publication: Hierarchical semi-Markov conditional random fields for deep recursive sequential data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2407880)