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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Due to the nature of deep learning approaches, it is inherently difficult tounderstand which aspects of a molecular graph drive the predictions of thenetwork. As a mitigation strategy, we constrain certain weights in a multi-taskgraph convolutional neural network according to the Gini index to maximize the"inequality" of the learned representations. We show that this constraint doesnot degrade evaluation metrics for some targets, and allows us to combine theoutputs of the graph convolutional operation in a visually interpretable way.We then perform a proof-of-concept experiment on quantum chemistry targets onthe public QM9 dataset, and a larger experiment on ADMET targets on proprietarydrug-like molecules. Since a benchmark of explainability in the latter case isdifficult, we informally surveyed medicinal chemists within our organization tocheck for agreement between regions of the molecule they and the modelidentified as relevant to the properties in question.

Gini in a Bottleneck: Sparse Molecular Representations for Graph Convolutional Neural Networks