BibTex format
@unpublished{Shanahan:2020,
author = {Shanahan, M and Nikiforou, K and Creswell, A and Kaplanis, C and Barrett, D and Garnelo, M},
title = {An explicitly relational neural network architecture},
year = {2020}
}
Publications
Citation
RIS format (EndNote, RefMan)
TY - UNPB
AB - With a view to bridging the gap between deep learning and symbolic AI, we present a novel endto-end neural network architecture that learns to form propositional representations with an explicitly relational structure from raw pixel data. In order to evaluate and analyse the architecture, we introduce a family of simple visual relational reasoning tasks of varying complexity. We show that the proposed architecture, when pre-trained on a curriculum of such tasks, learns to generate reusable representations that better facilitate subsequent learning on previously unseen tasks when compared to a number of baseline architectures. The workings of a successfully trained model are visualised to shed some light on how the architecture functions.
AU - Shanahan,M
AU - Nikiforou,K
AU - Creswell,A
AU - Kaplanis,C
AU - Barrett,D
AU - Garnelo,M
PY - 2020///
TI - An explicitly relational neural network architecture
ER -