Modular meta-learning in abstract graph networks for combinatorial generalization
Machine Learning
2018-12-20 v1 Machine Learning
Abstract
Modular meta-learning is a new framework that generalizes to unseen datasets by combining a small set of neural modules in different ways. In this work we propose abstract graph networks: using graphs as abstractions of a system's subparts without a fixed assignment of nodes to system subparts, for which we would need supervision. We combine this idea with modular meta-learning to get a flexible framework with combinatorial generalization to new tasks built in. We then use it to model the pushing of arbitrarily shaped objects from little or no training data.
Cite
@article{arxiv.1812.07768,
title = {Modular meta-learning in abstract graph networks for combinatorial generalization},
author = {Ferran Alet and Maria Bauza and Alberto Rodriguez and Tomas Lozano-Perez and Leslie P. Kaelbling},
journal= {arXiv preprint arXiv:1812.07768},
year = {2018}
}
Comments
Presented at NeurIPS meta-learning workshop 2018