Research into the potential of state-of-the-art machine learning solutions
The Saxon Institute for Computational Intelligence and Machine Learning (SICIM) aims to enable people from science and industry to recognize the potential of state-of-the-art machine learning solutions and to make the corresponding methods applicable.
At SICIM, various aspects of machine learning are researched both from a mathematical-theoretical and application-oriented perspective.
Theoretical research focuses on, among other things
- interpretable AI systems
- AI models with limited resources (smart systems AI)
- Decision reliability of AI classification systems
- Robustness of AI systems
Application-oriented problems include, among others
- Model confidence and rejection strategies
- transfer learning
- intelligent sensor fusion
Fields of application include
- Bioinformatics and medicine
- Diagnostic support
- Analysis of sequence data in molecular biology
- Spectral fingerprinting in agriculture
- Sensor data analysis and sensor fusion in technical systems
- Smart sensor systems
- Object recognition in autonomous driving
- Banknote detection
- AI for complex technical systems
- Motion detection and analysis of people
- Plant control