Learning Structural Kernels for Natural Language Processing
- Daniel Beck,
- Trevor Cohn,
- ,
- Lucia Specia
- University of Sheffield,
- University of Melbourne,
- Uppsala University
Research Output:
Journal Article or Conference Article in Journal
Journal article
Peer-reviewOpen access
Publication Information
Output type
Research Output:
Journal Article or Conference Article in Journal
Journal article
Peer-reviewOriginal language
EnglishPages from-to (Number of pages)
Pages 461-473 (13 pages)Journal (Volume, Issue Number)
Transactions of the Association for Computational Linguistics (Volume 3)Publication milestones
- Published - 2015
Publication status
Published - 2015
ISSN
2307-387XPublication IDs
- ORCID: /0000-0002-6103-7275/work/106363232
Abstract
Structural kernels are a flexible learning paradigm that has been widely used in Natural Language Processing. However, the problem of model selection in kernel-based methods is usually overlooked. Previous approaches mostly rely on setting default values for kernel hyperparameters or using grid search, which is slow and coarse-grained. In contrast, Bayesian methods allow efficient model selection by maximizing the evidence on the training data through gradient-based methods. In this paper we show how to perform this in the context of structural kernels by using Gaussian Processes. Experimental results on tree kernels show that this procedure results in better prediction performance compared to hyperparameter optimization via grid search. The framework proposed in this paper can be adapted to other structures besides trees, e.g., strings and graphs, thereby extending the utility of kernel-based methods.
Publication metrics
PlumX, opens in new tab
Captures
104
Citations
3
