Output details
11 - Computer Science and Informatics
University of Edinburgh
Article title
Greedy Learning of Binary Latent Trees
Type
D - Journal article
Title of journal
IEEE Transactions on Pattern Analysis and Machine Intelligence
Article number
-
Volume number
33
Issue number
6
First page of article
1087
ISSN of journal
0162-8828
Year of publication
2011
Number of additional authors
1
Additional information
<24> Originality: This paper introduced a greedy strategy for unsupervised learning of latent tree models from data. This is substantially faster than previous methods by Zhang, yet produces models of comparable quality. Latent tree models are important due to their interpretability, which can shed light on data generating processes.
Significance: An efficient algorithm for learning latent trees will enhance the take up of these methods in unsupervised learning. An implementation was published along with the paper.
Rigour: The paper appears in the leading pattern analysis journal IEEE PAMI -- 2010 impact factor of 5.027.
Interdisciplinary
-
Cross-referral requested
-
Research group
B - Institute for Adaptive & Neural Computation
Citation count
6
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-