DeepAI
Log In Sign Up

TexRel: a Green Family of Datasets for Emergent Communications on Relations

05/26/2021
by   Hugh Perkins, et al.
0

We propose a new dataset TexRel as a playground for the study of emergent communications, in particular for relations. By comparison with other relations datasets, TexRel provides rapid training and experimentation, whilst being sufficiently large to avoid overfitting in the context of emergent communications. By comparison with using symbolic inputs, TexRel provides a more realistic alternative whilst remaining efficient and fast to learn. We compare the performance of TexRel with a related relations dataset Shapeworld. We provide baseline performance results on TexRel for sender architectures, receiver architectures and end-to-end architectures. We examine the effect of multitask learning in the context of shapes, colors and relations on accuracy, topological similarity and clustering precision. We investigate whether increasing the size of the latent meaning space improves metrics of compositionality. We carry out a case-study on using TexRel to reproduce the results of an experiment in a recent paper that used symbolic inputs, but using our own non-symbolic inputs, from TexRel, instead.

READ FULL TEXT

page 1

page 2

page 3

page 4

06/14/2021

pix2rule: End-to-end Neuro-symbolic Rule Learning

Humans have the ability to seamlessly combine low-level visual input wit...
05/25/2022

Inductive Learning of Complex Knowledge from Raw Data

One of the ultimate goals of Artificial Intelligence is to learn general...
06/01/2021

Learning Representations for Sub-Symbolic Reasoning

Neuro-symbolic methods integrate neural architectures, knowledge represe...
05/20/2020

A refinement checking based strategy for component-based systems evolution

We propose inheritance and refinement relations for a CSP-based componen...
02/12/2019

Multitask Learning for Polyphonic Piano Transcription, a Case Study

Viewing polyphonic piano transcription as a multitask learning problem, ...
07/26/2000

Entrenchment Relations: A Uniform Approach to Nonmonotonicity

We show that Gabbay's nonmonotonic consequence relations can be reduced ...
09/28/2015

A Preliminary Study on the Learning Informativeness of Data Subsets

Estimating the internal state of a robotic system is complex: this is pe...

Code Repositories

texrel

Code for paper "TexRel: a Green Family of Datasets for Emergent Communications on Relations", Hugh Perkins, 2021 (https://arxiv.org/abs/2105.12804)


view repo