Transfer learning for domain adaptation in soybean yield prediction

• Grant number: 25/10840-3
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: August 01, 2025
• End date: July 31, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Zhao Liang
• Grantee: André Oliveira Françani
• Host Institution: Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (FFCLRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil
• Associated research grant: 24/15430-5 - Machine learning-based multi-modal data fusion and growth modeling for soybean production improvement, AP.PFPMCG.TEM

Abstract

Accurate and generalizable crop yield prediction remains a challenge in agriculture, especially when dealing with data from heterogeneous regions. This project proposes the development of deep learning models for soybean yield prediction by leveraging transfer learning and domain adaptation strategies to address geographic and environmental variability. By utilizing multispectral and hyperspectral sensor data, we aim to build robust models capable of transferring knowledge across different domains. The study integrates regularization techniques, unsupervised domain adaptation, as well as complementary deep learning approaches such as image augmentation, and explainable artificial intelligence techniques. We expect to develop accurate predictive models that can be applied to diverse agricultural conditions, supporting data-driven decision-making for sustainable crop production.