MINERVA
MINERVA is a research project developing a new generation of transparent and interpretable AI for visual reasoning. The project combines neural perception with symbolic reasoning to overcome a key limitation of conventional deep-learning systems: while they can recognize complex visual patterns, their internal decision processes are often difficult to understand, verify, or audit. MINERVA is designed to bridge this gap by creating a modular neurosymbolic AI architecture in which visual information can be represented, decomposed, and reasoned about in a structured and traceable way.
- Duration
- 01.01.2027 – 31.05.2029
- Doctoral student
- Sebastian Pritz
- Funding
- Austrian Research Promotion Agency FFG – Dissertationsprogramm der Fachhochschule OÖ