Several efforts have been developed recently to integrate (deep) Machine Learning (ML) algorithms into crop models to result in better predictions and inform adaptation strategies. In this session our speakers present the current work they are developing to link training genetic data to crop models to improve predictions and facilitate genomic selection for upstream crop management support. Some examples of image analysis and phenotyping using ML linked to crop models are presented. After the presentation, our panelists open up a discussion about the role of ML in crop modeling.

Speakers

• Matthew Reynolds: CIMMYT, Crop Physiologist
• Scott Chapman: The University of Queensland, Professor
• Mark Cooper: The University of Queensland QAAFI, Chair Crop Improvement