Learning Distributed Representations by Mapping Concepts and Relations into a Linear Space
Abstract
Linear Relational Embedding is a method of learning a distributed representation of concepts from data consisting of binary relations between concepts. Concepts are represented as vectors, binary relations as matrices, and the operation of applying a relation to a concept as a matrix-vector multiplication that produces an approximation to the related concept. A representation for concepts and relations is learned by maximizing an appropriate discriminative goodness function using gradient ascent. On a task involving family relationships, learning is fast and leads to good generalization.
Cite
Text
Paccanaro and Hinton. "Learning Distributed Representations by Mapping Concepts and Relations into a Linear Space." International Conference on Machine Learning, 2000.Markdown
[Paccanaro and Hinton. "Learning Distributed Representations by Mapping Concepts and Relations into a Linear Space." International Conference on Machine Learning, 2000.](https://mlanthology.org/icml/2000/paccanaro2000icml-learning/)BibTeX
@inproceedings{paccanaro2000icml-learning,
title = {{Learning Distributed Representations by Mapping Concepts and Relations into a Linear Space}},
author = {Paccanaro, Alberto and Hinton, Geoffrey E.},
booktitle = {International Conference on Machine Learning},
year = {2000},
pages = {711-718},
url = {https://mlanthology.org/icml/2000/paccanaro2000icml-learning/}
}