An asymmetric focal loss function improves graph neural network DDI prediction on TWOSIDES, increasing accuracy from 0.699 to 0.892 without architectural changes.
Generating Explainable Hypotheses for Drug Repurposing with Graph Neural Networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Asymmetric Focal Loss Improves Graph Neural Network Prediction of Drug-Drug Interactions
An asymmetric focal loss function improves graph neural network DDI prediction on TWOSIDES, increasing accuracy from 0.699 to 0.892 without architectural changes.