Chairs: Debaditya Chakraborty, Hakan BaÅŸaÄŸaoÄŸlu (University of Texas San Antonio, Edwards Aquifer Authority) | ![]() ![]() ![]() ![]() |
Explainable Artificial Intelligence (xAI) is emerging as a versatile tool to reveal data-driven novel information from the sheer volume of non-linearly related multidimensional data in agronomy, hydrology, and energy domains. There is a current need for further studies to enhance the coupling of explanatory methods with the AI models (leading to xAI models) to assess the importance of predictors in predicting the predictands, explain the inter-dependencies and interrelations between them, justify the xAI-based decisions, and explore new knowledge. To address these needs, this special track focuses on the application of xAI in Food Energy Water (FEW) domains and how it enables accurate predictions while revealing new knowledge in the respective domains. Some examples of topics of interest include but are not limited to:
Explanatory methods integrated with AI models in FEW |
Challenges and solutions of XAI models for short-term or long-term decisions in the FEW domains |
XAI-based assessment and prediction of climate change impacts on croplands, water resources, and energy production in FEW |
Predictions and analysis of physical behaviors using XAI in FEW domains |
XAI-based assessments and predictions of extreme events (e.g., droughts and floods) in the context of the assumption of stationarity |
Neural networks vs. Tree-based models for generating better predictions and explanations in FEW domains |