Multilabel structured output learning with random spanning trees of max-margin Markov networks
Marchand M, Hongyu S, Morvant E, Rousu J, Shawe Taylor J. 2014. Multilabel structured output learning with random spanning trees of max-margin Markov networks. Advances in Neural Information Processing Systems. NIPS: Neural Information Processing Systems vol. 27.
Download (ext.)
Conference Paper
| Published
| English
Author
Marchand, Mario;
Hongyu, Su;
Morvant, EmilieISTA 
;
Rousu, Juho;
Shawe Taylor, John
;
Rousu, Juho;
Shawe Taylor, JohnDepartment
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.
Publishing Year
Date Published
2014-01-01
Proceedings Title
Advances in Neural Information Processing Systems
Publisher
Neural Information Processing Systems Foundation
Volume
27
Conference
NIPS: Neural Information Processing Systems
Conference Location
Montreal, Canada
Conference Date
2014-12-08 – 2014-12-13
ISBN
IST-REx-ID
Cite this
Marchand M, Hongyu S, Morvant E, Rousu J, Shawe Taylor J. Multilabel structured output learning with random spanning trees of max-margin Markov networks. In: Advances in Neural Information Processing Systems. Vol 27. Neural Information Processing Systems Foundation; 2014.
Marchand, M., Hongyu, S., Morvant, E., Rousu, J., & Shawe Taylor, J. (2014). Multilabel structured output learning with random spanning trees of max-margin Markov networks. In Advances in Neural Information Processing Systems (Vol. 27). Montreal, Canada: Neural Information Processing Systems Foundation.
Marchand, Mario, Su Hongyu, Emilie Morvant, Juho Rousu, and John Shawe Taylor. “Multilabel Structured Output Learning with Random Spanning Trees of Max-Margin Markov Networks.” In Advances in Neural Information Processing Systems, Vol. 27. Neural Information Processing Systems Foundation, 2014.
M. Marchand, S. Hongyu, E. Morvant, J. Rousu, and J. Shawe Taylor, “Multilabel structured output learning with random spanning trees of max-margin Markov networks,” in Advances in Neural Information Processing Systems, Montreal, Canada, 2014, vol. 27.
Marchand M, Hongyu S, Morvant E, Rousu J, Shawe Taylor J. 2014. Multilabel structured output learning with random spanning trees of max-margin Markov networks. Advances in Neural Information Processing Systems. NIPS: Neural Information Processing Systems vol. 27.
Marchand, Mario, et al. “Multilabel Structured Output Learning with Random Spanning Trees of Max-Margin Markov Networks.” Advances in Neural Information Processing Systems, vol. 27, Neural Information Processing Systems Foundation, 2014.
All files available under the following license(s):
Copyright Statement:
This Item is protected by copyright and/or related rights. [...]
Link(s) to Main File(s)
Access Level
Open Access
Google Scholar
ISBN Search