A harmonic feature mapping plus a five-term 'chemically-informed' loss modestly improves F1 for molecular odor prediction on 5,788 molecules, with no statistical validation.
Smiles to smell: decoding the structure–odor relationship of chemical compounds using the deep neural network approach,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Molecular Odor Prediction with Harmonic Modulated Feature Mapping and Chemically-Informed Loss
A harmonic feature mapping plus a five-term 'chemically-informed' loss modestly improves F1 for molecular odor prediction on 5,788 molecules, with no statistical validation.