The entity responsible: Academic Computer Centre CYFRONET AGH University of Science and Technology (ACC Cyfronet AGH)
The main purpose of the task is to develop software to support the planning and conduction of research on IB EPOS-PL data using artificial intelligence methods (primarily various machine learning methods). The established platform will enable the easy use of data integrated in the EPOS program (including IB EPOS-PL), the selection of machine learning methods and the execution of resource-intensive stages of processing using the analytical assets of the Polish High-Performance Computer Centres.
The result of the development work will be the establishment of the unique EPOS-AI platform. The platform will encompass access to the collected data, the analytical assets, and the necessary software enabling the execution of tasks involving artificial intelligence methods by means of supercomputer infrastructure. The proposed EPOS-AI platform is a unique and dedicated AI for the EPOS work environment. The algorithms serving as the foundation for the use of artificial intelligence methods will be based on the specialised knowledge and experience of researchers belonging to the consortium members cooperating on the establishment of the platform. Verification of the developed algorithms, AI training and assessment of the AI’s performance will be possible thanks to the unique competence and expertise of the aforementioned researchers. As far as this is concerned, only the Consortium employees possess the sufficient specialist knowledge necessary to establish the EPOS-AI platform.