DEEC Talk: “Knowledge‑Centric AI for Scientific Discovery”

On December 17, the session “Knowledge‑Centric AI for Scientific Discovery” took place, featuring Carla P. Gomes, Ronald C. and Antonia V. Nielsen Professor of Computer Science and Information Science at Cornell University, director of the Institute for Computational Sustainability, and co-director of the Cornell AI for Science Institute.

The researcher discussed some of the latest applications of artificial intelligence that contribute to sustainable development, highlighting their impact not only on the environment but also on social and economic aspects.
Carla P. Gomes began by emphasizing that artificial intelligence is no longer confined to academia and is increasingly affecting society through various applications, such as goods management in factory environments and the generation of text and images through language models like ChatGPT. Regarding sustainability, AI can be used for biodiversity protection—for example, studying the distribution of species such as migratory birds—as well as for resource management and planning.

Her research integrates data processing, allowing explicit analysis of different contexts and generating interpretable recommendations, including in applications of computer vision, territorial mapping, and the study of materials for renewable energy systems.
During her presentation, Carla P. Gomes drew a parallel between her research and the organization of subway lines, where each route can be seen as a component of the system, encompassing probabilistic inference and optimization (computational methods for predicting and making decisions about complex systems), sequential decision-making, citizen science contributions, the temporal and spatial scale of models, and pattern separation. The "stations" represent the application areas of these systems.

Carla P. Gomes also highlighted the importance of research not relying solely on “black-box” systems, emphasizing the need to study the mechanisms underlying AI decision-making tools. She concluded her talk by noting that system development is an ongoing process, continuously accompanied by emerging challenges, ideas, and methodologies over time.

