Combining expert knowledge and neural networks to model environmental stresses in agriculture

TL;DR

This paper integrates expert agricultural knowledge with neural networks to model environmental stresses like heat and drought, aiming to improve prediction and resilience assessment.

Contribution

It introduces a hybrid approach combining expert models with neural networks and performs sensitivity analysis for stress susceptibility classification.

Findings

Expert models serve as benchmarks for neural network design.

Neural networks can be clustered into susceptible and resistant groups.

Hybrid models improve understanding of environmental stress impacts.

Abstract

In this work we combine representation learning capabilities of neural network with agricultural knowledge from experts to model environmental heat and drought stresses. We first design deterministic expert models which serve as a benchmark and inform the design of flexible neural-network architectures. Finally, a sensitivity analysis of the latter allows a clustering of hybrids into susceptible and resistant ones.