2024/05/29 by Boje Deforce, Deforce, Boje, Bart Baesens +3 · 3 citations
Agricultural and Biological Sciences · Engineering · #Agricultural engineering #Agriculture #Archaeology #Computer science #Engineering #Environmental science #FOS: Computer and information sciences #Foundation (evidence) #Geography #Geology #Geotechnical engineering #Machine Learning (cs.LG) #Machine learning #Meteorology #Moisture #Series (stratigraphy) #Smart Agriculture and AI #Time series #Water content
paper · pdf · doi:10.48550/arxiv.2405.18913
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2024/05/29 · openalex created_date 2024/05/31 · openalex updated_date 2026/07/28
The recent surge in foundation models for natural language processing and computer vision has fueled innovation across various domains. Inspired by this progress, we explore the potential of foundation models for time-series forecasting in smart agriculture, a field often plagued by limited data availability. Specifically, this work presents a novel application of TimeGPT, a state-of-the-art (SOTA) time-series foundation model, to predict soil water potential (ψsoil), a key indicator of field water status that is typically used for irrigation advice. Traditionally, this task relies on a wide array of input variables. We explore ψsoil's ability to forecast ψsoil in: (i) a zero-shot setting, (ii) a fine-tuned setting relying solely on historic ψsoil measurements, and (iii) a fine-tuned setting where we also add exogenous variables to the model. We compare TimeGPT's performance to established SOTA baseline models for forecasting ψsoil. Our results demonstrate that TimeGPT achieves competitive forecasting accuracy using only historical ψsoil data, highlighting its remarkable potential for agricultural applications. This research paves the way for foundation time-series models for sustainable development in agriculture by enabling forecasting tasks that were traditionally reliant on extensive data collection and domain expertise.