Multi-head attention neural network predicts grapevine leaf reflectance from 16 traits with average R² 0.84 and NRMSE 1.52%, showing lower MAE than PROSPECT-PRO especially in NIR and SWIR.
Guided inte- grated gradients: An adaptive path method for removing noise
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Enforcing feature decorrelation during training produces sharper saliency maps and higher accuracy on image classification benchmarks.
citing papers explorer
-
Leaf Spectral Reflectance Prediction Using Multi-Head Attention Neural Networks
Multi-head attention neural network predicts grapevine leaf reflectance from 16 traits with average R² 0.84 and NRMSE 1.52%, showing lower MAE than PROSPECT-PRO especially in NIR and SWIR.
-
SaliencyDecor: Enhancing Neural Network Interpretability through Feature Decorrelation
Enforcing feature decorrelation during training produces sharper saliency maps and higher accuracy on image classification benchmarks.