Convex co-embedding
- Farzaneh Mirzazadeh, University of Alberta
- Yuhong Guo
- Dale Schuurmans, AICML
We present a general framework for association learning, where entities are embedded in a common latent space to express relatedness by geometry -- an approach that underlies the state of the art for link prediction, relation learning, multi-label tagging, relevance retrieval and ranking. Although current approaches rely on local training applied to non-convex formulations, we demonstrate how general convex formulations can be achieved for entity embedding, both for standard multi-linear and prototype-distance models. We investigate an efficient optimization strategy that allows scaling. An experimental evaluation reveals the advantages of global training in different case studies.
Citation
F. Mirzazadeh, Y. Guo, D. Schuurmans. "Convex co-embedding". National Conference on Artificial Intelligence (AAAI), pp 1989-1996, July 2014.| Keywords: | Convex, Co-embedding |
| Category: | In Conference |
| Web Links: | AAAI |
BibTeX
@incollection{Mirzazadeh+al:AAAI14,
author = {Farzaneh Mirzazadeh and Yuhong Guo and Dale Schuurmans},
title = {Convex co-embedding},
Pages = {1989-1996},
booktitle = {National Conference on Artificial Intelligence (AAAI)},
year = 2014,
}Last Updated: February 19, 2020Submitted by Sabina P
