Multilabel structured output learning with random spanning trees of max-margin Markov networks Conference Paper


Author(s): Marchand, Mario; Hongyu, Su; Morvant, Emilie; Rousu, Juho; Shawe-Taylor, John
Title: Multilabel structured output learning with random spanning trees of max-margin Markov networks
Affiliation
Abstract: We show that the usual score function for conditional Markov networks can be written as the expectation over the scores of their spanning trees. We also show that a small random sample of these output trees can attain a significant fraction of the margin obtained by the complete graph and we provide conditions under which we can perform tractable inference. The experimental results confirm that practical learning is scalable to realistic datasets using this approach.
Conference Title: NIPS: Neural Information Processing Systems
Conference Dates: December 8-12, 2014
Conference Location: Montréal, Canada
ISBN: 10495258
Publisher: Neural Information Processing Systems  
Date Published: 2014-01-01
URL:
Open access: yes (repository)