We observed that generally the embedding representation is very rich and information dense. For example, reducing the dimensionality of the inputs using SVD or PCA, even by 10%, generally results in worse downstream performance on specific tasks.
/**
* Called when a null model is about to be retrieved in order to allow a subclass to provide an
* initial model.
* <p>
* By default this implementation looks components in the parent chain owning a
* {@link IComponentInheritedModel} to provide a model for this component via
* {@link IComponentInheritedModel#wrapOnInheritance(Component)}.
* <p>
* For example a {@link FormComponent} has the opportunity to instantiate a model on the fly
* using its {@code id} and the containing {@link Form}'s model, if the form holds a
* {@link CompoundPropertyModel}.
*
* @return The model
*/
protected IModel<?> initModel()
{
IModel<?> foundModel = null;
// Search parents for IComponentInheritedModel (i.e. CompoundPropertyModel)
for (Component current = getParent(); current != null; current = current.getParent())
{
// Get model
// Don't call the getModel() that could initialize many in between
// completely useless models.
// IModel model = current.getDefaultModel();
IModel<?> model = current.getModelImpl();
if (model instanceof IWrapModel && !(model instanceof IComponentInheritedModel))
{
model = ((IWrapModel<?>)model).getWrappedModel();
}
if (model instanceof IComponentInheritedModel)
{
// return the shared inherited
foundModel = ((IComponentInheritedModel<?>)model).wrapOnInheritance(this);
setFlag(FLAG_INHERITABLE_MODEL, true);
break;
}
}
// No model for this component!
return foundModel;
}
M. Schuhmacher, and S. Ponzetto. Proceedings of the 7th ACM International Conference on Web Search and Data Mining
, page 543--552. New York, NY, USA, ACM, (2014)
E. Alfonseca, K. Filippova, J. Delort, and G. Garrido. Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Short Papers - Volume 2
, page 54--59. Stroudsburg, PA, USA, Association for Computational Linguistics, (2012)
J. Kang, K. Lerman, and L. Getoor. (2013)cite arxiv:1301.6277Comment: The 2013 International Conference on Social Computing, Behavioral-Cultural Modeling, & Prediction (SBP 2013).
W. Headden, III, M. Johnson, and D. McClosky. Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics
, page 101--109. Stroudsburg, PA, USA, Association for Computational Linguistics, (2009)
J. Nivre. Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing
, page 351-359. Association for Computational Linguistics Morristown, NJ, USA, Association for Computational Linguistics Morristown, NJ, USA, (2009)Uppsala University.
D. Klein, and C. Manning. Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
, page 128--135. Stroudsburg, PA, USA, Association for Computational Linguistics, (2002)
P. Mika. The Semantic Web - ISWC 2005, Proceedings of the 4th International Semantic Web Conference, ISWC 2005, Galway, Ireland, November 6-10
, volume 3729 of Lecture Notes in Computer Science, page 522-536. Springer, (2005)