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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Graph neural networks (GNNs) have been proposed for a wide range ofgraph-related learning tasks. In particular, in recent years, an increasingnumber of GNN systems were applied to predict molecular properties. However, adirect impediment is to select appropriate hyperparameters to achievesatisfactory performance with lower computational cost. Meanwhile, manymolecular datasets are far smaller than many other datasets in typical deeplearning applications. Most hyperparameter optimization (HPO) methods have notbeen explored in terms of their efficiencies on such small datasets in themolecular domain. In this paper, we conducted a theoretical analysis of commonand specific features for two state-of-the-art and popular algorithms for HPO:TPE and CMA-ES, and we compared them with random search (RS), which is used asa baseline. Experimental studies are carried out on several benchmarks inMoleculeNet, from different perspectives to investigate the impact of RS, TPE,and CMA-ES on HPO of GNNs for molecular property prediction. In ourexperiments, we concluded that RS, TPE, and CMA-ES have their individualadvantages in tackling different specific molecular problems. Finally, webelieve our work will motivate further research on GNN as applied to molecularmachine learning problems in chemistry and materials sciences.

A Systematic Comparison Study on Hyperparameter Optimisation of Graph Neural Networks for Molecular Property Prediction