Sensitivity Analysis of Decision Rules in an Agent-Based Agro-Ecological Model
Not provided in the abstract
Abstract
The paper discusses the challenges of sensitivity analysis in agent-based agro-ecological models and proposes a novel method to analyze decision trees in this context.
Reality Card
The authors developed a new method, random-forest-supervised HSIC-ANOVA, to effectively analyze hierarchical decision spaces in agro-ecological models, providing interpretable insights into decision impacts.
The proposed method ranks main effects and interactions within a decision tree structure, improving the understanding of simulation outputs like nitrogen leaching and yield predictions.
The paper does not specify the reproducibility of the method across different agro-ecological models or datasets.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.