Deep Learning Meets Process-Based Models: A Hybrid Approach to Agricultural Challenges

2025-04-22

AI Summary

Process-based models (PBMs) in agricultural modeling suffer from poor scalability and parameterization challenges, while deep learning (DL) models exhibit limited interpretability and are prone to overfitting under small-sample conditions. Method: This paper proposes a bidirectionally collaborative hybrid modeling paradigm integrating physical mechanisms and data-driven learning. It innovatively constructs two complementary hybrid architectures—DL-informed and PBM-informed models—by embedding physics-based constraints and jointly optimizing shared parameters to enable mechanism-guided data learning and data-informed mechanistic calibration. Results: Evaluated on crop dry matter prediction, the hybrid models reduce mean absolute error by 23.7% relative to standalone PBMs and DL models, improve noise robustness by 41%, and enhance cross-site generalization accuracy by 35.2%. The approach significantly boosts model reliability and interpretability under small-sample, high-noise, and unseen-location scenarios.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Process-based models (PBMs) and deep learning (DL) are two key approaches in agricultural modelling, each offering distinct advantages and limitations. PBMs provide mechanistic insights based on physical and biological principles, ensuring interpretability and scientific rigour. However, they often struggle with scalability, parameterisation, and adaptation to heterogeneous environments. In contrast, DL models excel at capturing complex, nonlinear patterns from large datasets but may suffer from limited interpretability, high computational demands, and overfitting in data-scarce scenarios. This study presents a systematic review of PBMs, DL models, and hybrid PBM-DL frameworks, highlighting their applications in agricultural and environmental modelling. We classify hybrid PBM-DL approaches into DL-informed PBMs, where neural networks refine process-based models, and PBM-informed DL, where physical constraints guide deep learning predictions. Additionally, we conduct a case study on crop dry biomass prediction, comparing hybrid models against standalone PBMs and DL models under varying data quality, sample sizes, and spatial conditions. The results demonstrate that hybrid models consistently outperform traditional PBMs and DL models, offering greater robustness to noisy data and improved generalisation across unseen locations. Finally, we discuss key challenges, including model interpretability, scalability, and data requirements, alongside actionable recommendations for advancing hybrid modelling in agriculture. By integrating domain knowledge with AI-driven approaches, this study contributes to the development of scalable, interpretable, and reproducible agricultural models that support data-driven decision-making for sustainable agriculture.

Problem

• Combining PBMs and DL to overcome agricultural modeling limitations
• Evaluating hybrid PBM-DL models for crop biomass prediction
• Addressing interpretability and scalability in agricultural AI models

Innovation

• Hybrid PBM-DL approach combines mechanistic and data-driven insights
• DL-informed PBMs refine models with neural networks
• PBM-informed DL uses physical constraints to guide predictions