Optit is part of the TUPLES project on Artificial Intelligence, funded by the European Union through the HORIZON-CL4-2021-HUMAN-01 program. The project has the ambitious goal of bringing substantial evolution to the field of AI by creating scalable, transparent, robust, and secure algorithmic solutions for Program & Scheduling (P&S). By combining symbolic P&S methods with data-driven methods, the project seeks to leverage the strengths of each, ensuring efficiency, adaptability, and transparency.
The cornerstone lies in developing rigorous explanations and verification approaches, fundamental for transparency, robustness, and safety in machine-learned decision sequences. TUPLES intends to showcase these advancements through practical use cases, including production, aerial operations, sports management, waste collection, and energy management. The latter two use cases are managed by Optit, which also has the responsibility for the overall project communication.
The ultimate goal is to increase trust in AI-managed application solutions, promoting their adoption in various sectors. The team comprising this project includes industrial partners and universities with extensive experience in the field of artificial intelligence. The coexistence of these two aspects, one more theoretical and the other more grounded in the reality of real applications, aims to achieve concrete results rather than just excellent but impractical theoretical elaborations.
Among the industrial partners, in addition to Optit, there are Airbus and Scisports, while the academic partners include UNIBO, ANITI, CVUT, KU Leuven, and Saarland University. Although the teams’ work is regularly conducted remotely due to the various partners being located throughout Europe, it is nonetheless beneficial and valuable to meet in person. After meetings in Toulouse and Bologna, this time we gathered in the historic university city of Leuven in Belgium, at the DTAI headquarters, the Declarative Languages and Artificial Intelligence section, at the Department of Computer Science of KU Leuven.
For details and updates on the TUPLES project, visit the website www.tuples.ai.
