用多任务学习提升葡萄物候预测精度，兼顾生物机制与数据稀疏性问题。
Calibrating Biophysical Models for Grape Phenology Prediction via Multi-Task Learning
- 融合循环神经网络与多任务学习，参数化可微分的生物物理模型。
- 在真实与合成数据上，对物候阶段预测准确率显著优于传统模型和纯深度学习方法。
- 适合需要精细管理的葡萄种植者及农业智能化研究者使用。
精准预测葡萄物候对及时制定葡萄园管理决策（如灌溉、施肥）至关重要，有助于最大化产量与品质。尽管基于历史田间数据校准的传统生物物理模型可实现季节性预测，但其精度不足以支持精细化管理。深度学习虽具潜力，但受制于稀疏的物候数据，尤其在品种层面表现受限。本文提出一种混合建模方法，结合多任务学习与循环神经网络，对可微分生物物理模型进行参数化。通过多任务学习共享不同品种间的参数信息，同时保持生物结构，提升了预测的鲁棒性与准确性。基于真实世界与合成数据的实证评估表明，该方法在预测物候阶段及其他作物状态变量（如抗寒性、小麦产量）方面，显著优于传统生物物理模型与基准深度学习方法。
原文摘要 · Abstract (English)
Accurate prediction of grape phenology is essential for timely vineyard management decisions, such as scheduling irrigation and fertilization, to maximize crop yield and quality. While traditional biophysical models calibrated on historical field data can be used for season-long predictions, they lack the precision required for fine-grained vineyard management. Deep learning methods are a compelling alternative but their performance is hindered by sparse phenology datasets, particularly at the cultivar level. We propose a hybrid modeling approach that combines multi-task learning with a recurrent neural network to parameterize a differentiable biophysical model. By using multi-task learning to predict the parameters of the biophysical model, our approach enables shared learning across cultivars while preserving biological structure, thereby improving the robustness and accuracy of predictions. Empirical evaluation using real-world and synthetic datasets demonstrates that our method significantly outperforms both conventional biophysical models and baseline deep learning approaches in predicting phenological stages, as well as other crop state variables such as cold-hardiness and wheat yield.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。