- Published on
Relationships from Entity Stream
- Authors
- Name
- Martin Andrews
- @mdda123
This paper was accepted to the ViGIL workshop at NIPS-2017 in Long Beach, California, USA.
Abstract
Relational reasoning is a central component of intelligent behavior, but has proven difficult for neural networks to learn. The Relation Network (RN) module was recently proposed by DeepMind to solve such problems, and demonstrated state-of-the-art results on a number of datasets. However, the RN module scales quadratically in the size of the input, since it calculates relationship factors between every patch in the visual field, including those that do not correspond to entities. In this paper, we describe an architecture that enables relationships to be determined from a stream of entities obtained by an attention mechanism over the input field. The model is trained end-to-end, and demonstrates equivalent performance with greater interpretability while requiring only a fraction of the model parameters of the original RN module.
Poster Version
Link to Paper
And the BiBTeX entry for the arXiv version:
@misc{andrews2019relationships,
title={Relationships from Entity Stream},
author={Martin Andrews and Sam Witteveen},
year={2019},
eprint={1909.03315},
archivePrefix={arXiv},
primaryClass={cs.CL}
}