Marc Semper Lloret — Publications https://marcsemperlloret.com/publications/ Researcher in reliable spatiotemporal artificial intelligence, graph neural networks, sensor data quality and environmental forecasting. en A practical decision-support system for robust post-training model selection in spatiotemporal forecasting https://marcsemperlloret.com/publications/robust-post-training-model-selection/ https://marcsemperlloret.com/publications/robust-post-training-model-selection/ Thu, 23 Jul 2026 12:00:00 GMT Choosing the best forecasting model from one clean validation set can produce a fragile deployment decision. This work tests whether that decision survives realistic changes to the evaluation reference and turns the diagnosis into an auditable recommendation. Multi-Dataset Training for Improved Accuracy in Spatio-Temporal Problems: An Explainable Analysis https://marcsemperlloret.com/publications/multi-dataset-training-spatiotemporal/ https://marcsemperlloret.com/publications/multi-dataset-training-spatiotemporal/ Sat, 07 Mar 2026 12:00:00 GMT This study asks whether a graph forecasting model can learn better node representations by training across several related datasets instead of learning every representation from scratch. Global forecasting of aerosol optical depth through a deep learning spatiotemporal modeling https://marcsemperlloret.com/publications/global-aerosol-optical-depth/ https://marcsemperlloret.com/publications/global-aerosol-optical-depth/ Mon, 01 Dec 2025 12:00:00 GMT The work models how atmospheric aerosols evolve across the planet by combining a global graph of observations with temporal patterns at several scales. Spatio-temporal graph neural network for inter-city air quality forecasting https://marcsemperlloret.com/publications/inter-city-air-quality/ https://marcsemperlloret.com/publications/inter-city-air-quality/ Fri, 28 Nov 2025 12:00:00 GMT Air quality at one monitoring station is connected to conditions elsewhere. This work learns both nearby and long-range relationships across the Spanish monitoring network. Noise Pollution Prediction in a Densely Populated City Using a Spatio-Temporal Deep Learning Approach https://marcsemperlloret.com/publications/noise-pollution-madrid/ https://marcsemperlloret.com/publications/noise-pollution-madrid/ Fri, 16 May 2025 12:00:00 GMT Urban noise has both a temporal rhythm and a spatial structure. The study compares deep-learning approaches that represent those two dimensions explicitly across Madrid. Global forecasting of carbon concentration through a deep learning spatiotemporal modeling https://marcsemperlloret.com/publications/global-carbon-concentration/ https://marcsemperlloret.com/publications/global-carbon-concentration/ Fri, 15 Nov 2024 12:00:00 GMT This research tests deep-learning strategies for forecasting global carbon dioxide and methane concentrations six months ahead from satellite and environmental data.