arXiv cs.CY· Abdelghani Belgaid·· 11 小时前AI 评分19
基于机器学习的定点施肥推荐中的算法单一化问题
Algorithmic Monoculture in Machine Learning-based Site-Specific Fertilizer Recommendation
AI 导读
研究指出机器学习施肥推荐常按单模型评估，但多农场部署时相关误差会聚合为系统性成本。基于摩洛哥五类谷类系统九个模型族的基准校准显示，在10% regret 上限下，精度约束下的多样化使五个系统的建模社会损失降低 10.89%-41.53%（均值 19.41%），而均匀分配虽降低损失却在每种作物上都突破 regret 上限。
正文 · 原文
Abstract:Machine learning recommendations are often evaluated one model at a time, despite deployment across interconnected decision makers. For continuous NPK fertilizer prescriptions, correlated errors can aggregate across farms and generate systemic costs. A model of shared and farm-specific recommendation errors links expected social loss to model adoption shares; the Herfindahl-Hirschman Index characterizes concentration effects when common model errors are independent and predictive risks are equal. A covariance-generalized formulation captures the dependence between model families. A two-model threshold characterizes when privately optimal monoculture is socially dominated, and a convex portfolio rule limits average private predictive regret. Nine model families across five Moroccan cereal systems provide a benchmark-based calibration. Under illustrative parameters and a 10% regret cap, accuracy-constrained diversification lowers modeled social loss in all five systems by 10.89-41.53% (mean 19.41%). Uniform allocation reduces modeled loss in this setting, but breaches the regret cap in every crop. Correlation sensitivity shows that highly dependent model errors attenuate the benefits of diversification. These counterfactual outcomes are not estimates of agronomic or monetary welfare; the framework identifies adoption architecture and error dependence as ecosystem-level evaluation variables.
| Comments: | Accepted at the NeurIPS 2026 Workshop on Economics for Machine Learning (EconML) |
| Subjects: | Machine Learning (cs.LG); Computers and Society (cs.CY); Theoretical Economics (econ.TH) |
| Cite as: | arXiv:2610.36941 [cs.LG] |
| (or arXiv:2610.36941v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36941 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Abdelghani Belgaid [view email]
[v1]
Tue, 27 Oct 2026 10:01:56 UTC (115 KB)
来源：arXiv cs.CY · arxiv.org