EDIH Czech Technical University in Prague
Program
Projekty podpořené ze zahraničí (pracovní kód k dodatečnému upřesnění)
Provider
Another foreign provider
Investigators
Period
2023 - 2025
Description
The EDIH CTU represents a major European Digital Innovation Hub in the Czech Republic in the field of Artificial Intelligence (AI) and Machine Learning (ML) transferring trustworthy solutions and services to the industry, health, transportation and energy sectors. The EDIH CTU, with its vision "Inspire & Make the Czech AI-driven Industry", aims to become the innovation and technology leader in providing professional AI/ML services for the local SMEs, small mid-caps and public sector organizations with respect to their digital and green transformation. The strong consortium partners representing academia, business sector and key associations of enterprises in the Czech Republic provide sufficient know-how, expertise and state-of-the-art experimental facilities to serve the needs of the public and private sector at the national and also international level. The major focus is on promoting digital transformation adoption, providing high-quality services, education and knowledge sharing, pursuing ecosystem development, and establishing strong inter-EDIH collaboration. The consortium builds upon already existing partnerships in AI and manufacturing enabling best practice sharing, expertise exchange, and joint activities seeking far-reaching synergies thus strengthening the far-reaching impacts of the European network of EDIHs.
Symbolic and Subsymbolic Algorithmic Techniques for Modern Artificial Intelligence
Program
Studentská grantová soutěž ČVUT
Departments
Investigators
Period
2026 - 2028
Description
The project is aimed at supporting the research of doctoral students at the Department of Applied Mathematics at the Faculty of Information Technology. The topics that the doctoral students are working on include symbolic artificial intelligence techniques based on reasoning in models of the world, but also subsymbolic techniques that synthesize reasoning from training data. Within the project, we would like to connect these two paradigms as much as possible, on the symbolic side to use more available data to create models, and on the subsymbolic side not to rely entirely on machine learning, but also to address the explainability and safety of the proposed techniques.