English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning has consistently defied state-of-the-art techniques in manyfields over the last decade. However, we are just beginning to understand thecapabilities of neural learning in symbolic domains. Deep learningarchitectures that employ parameter sharing over graphs can produce modelswhich can be trained on complex properties of relational data. These includehighly relevant NP-Complete problems, such as SAT and TSP. In this work, weshowcase how Graph Neural Networks (GNN) can be engineered -- with a verysimple architecture -- to solve the fundamental combinatorial problem of graphcolouring. Our results show that the model, which achieves high accuracy upontraining on random instances, is able to generalise to graph distributionsdifferent from those seen at training time. Further, it performs better thanthe Neurosat, Tabucol and greedy baselines for some distributions. In addition,we show how vertex embeddings can be clustered in multidimensional spaces toyield constructive solutions even though our model is only trained as a binaryclassifier. In summary, our results contribute to shorten the gap in ourunderstanding of the algorithms learned by GNNs, as well as hoarding empiricalevidence for their capability on hard combinatorial problems. Our results thuscontribute to the standing challenge of integrating robust learning andsymbolic reasoning in Deep Learning systems.

Graph Colouring Meets Deep Learning: Effective Graph Neural Network Models for Combinatorial Problems