Global AI for Agriculture Modelling Community Begins in Wageningen
In a groundbreaking gathering in late January in Wageningen, researchers and experts from around the world celebrated the launch of the first-ever AgML workshop. This workshop aims to advance the use of machine learning for agricultural modelling and marks the beginning of an important international collaboration.
Agricultural models play a crucial role in understanding the impacts of climate change on the global food system and enhancing resilience. Machine learning methods offer new opportunities in scientific research by allowing the identification and analysis of complex relationships from large volumes of data, thereby improving agricultural modelling practices significantly.
To better understand the benefits and challenges of machine learning methods in agricultural modelling, the Agricultural Model Intercomparison and Improvement Project (AgMIP) has made coordinated efforts over the past year. As a result, AgML has emerged as a research initiative of AgMIP, a transdisciplinary community aimed at exploring the potential benefits and pitfalls of machine learning methods in agricultural modelling tasks.
The objectives of AgML include knowledge sharing, promotion of best practices for using machine learning tools in agricultural modelling, quantifying machine learning performance in crop modelling, and developing new methods tailored to the unique challenges faced by the agricultural sector.
Led by Professor Ioannis Athanasiadis, an expert in Artificial Intelligence and Data Science at WUR, and Lily Belle Sweet, a PhD candidate at UFZ, the workshop gathered representatives from organizations such as FAO, EU JRC, CGIAR, NASA, and numerous universities. Participants engaged in discussions on new machine learning paradigms and methodologies and emphasized the need for rigorous model evaluation.
In addition to discussions, the participants also took part in a hackathon where they built their own machine learning models. These models will be analyzed in AgML’s model comparison experiments, which aim to provide reliable estimates of machine learning architectures and methodologies for use in agricultural modelling applications.
AgML also focuses on creating open datasets to evaluate machine learning models in predicting climate impacts on agriculture and subnational yield forecasts for various crops and regions worldwide.
The collaborative effort of AgML brings together experts from various fields to develop AI benchmarks for reproducible, comparable, and interpretable modelling of agricultural and food systems.
This international collaboration facilitated by the AgMIP network welcomes all interested parties, including students, academics, and the private sector.
In conclusion, the launch of the AgML workshop sets the stage for a global initiative in using machine learning for agricultural modelling. The workshop not only facilitates knowledge sharing but also aims to address the unique challenges in the agricultural sector and develop solutions for a more sustainable and resilient food system.
References:
– [Global AI for Agriculture Modelling Community Begins in Wageningen](http://www.miragenews.com/global-ai-for-agriculture-modelling-community-563189/)
– [AgML Workshop](https://www.wur.nl/en/newsarticle/AgML-Workshop.htm)
