Home

GVA PROMETEO PROJECT - AI FOR COMPLEX SYSTEMS

This workshop aims to bring together researchers interested AI for complex systems, with especially attention to the Earth, Visual Brain and Social Systems. We will explore connections and similarities, and sysnergistically exploit AI methods for modeling and understanding: from advanced sensory systems, to foundation and deep learning models, hybrid modeling and causality

Explore the ISP Group

Our research spans various fields including statistical learning for Earth system modeling, human vision understanding, machine learning for remote sensing, and more. We have experts in areas such as signal processing, kernel methods, image statistics, and causal inference, all contributing to advancements in AI and complex systems.

Learn more

Workshop – AI for Complex Systems: Brain, Earth, Climate, Society

What: A 2-day workshop bringing together machine learning and visual neuroscience experts to discuss new methods and common challenges in addressing complexity in the Earth and climate, visual brain, and social systems.
Where: AIDET (room 1.1 & 1.2)
When: 13th - 15th January 2025

Why this workshop?

In AI4CS we develop advanced AI methods to model and understand complex systems, focusing on the visual brain, Earth and climate systems, and biosphere-anthroposphere interactions. A perfect storm is over us:

An ever-increasing amount of observational and sensory data.
Improved high-resolution yet mechanistic models are available.
Advanced machine learning techniques able to extract patterns and identify drivers from data.

In the last decade, machine learning models have helped to monitor, predict, and forecast all kinds of variables and parameters of interest from observational data. They help quantify visual stimuli, monitor land, oceans, and the atmosphere, and study socio-economic variables at different scales and spheres.

Current approaches, however, face three important challenges:

They cannot deal efficiently with the particular characteristics of data.
They do not respect the most elementary laws of physics.
They just interpolate but nothing fundamental is learned from data.

AI4CS aims to address these challenges with innovative, physics-aware, and causality-driven AI solutions to advance our understanding of these interconnected systems.

This workshop will address these goals, bringing together researchers to share progress, brainstorm new ideas, and foster interdisciplinary collaboration.