SS 02 - Challenges of Machine Learning in Intelligent Technical Systems

Special Session Organized by

Diana Göhringer, Technische Universität Dresden, Germany and Christoph-Alexander Holst, inIT – Institute Industrial IT, Germany and Anton Pfeifer, inIT – Institute Industrial IT, Germany

Download Call for Papers

Click here to download the session cfp.

Focus

Machine Learning techniques have achieved outstanding performances in numerous computing problems. But the requirements of technical systems still pose considerable challenges for data-driven machine learning methods. Challenges lie in the nature of available data resulting in uncertainties, conflict, incompleteness, low quality, and sparsity. On the other hand, technical systems require learners and models which are able to run in real-time, on limited hardware, are interpretable, robust, highly accurate, adaptable, and secure. This special session aims at advancing and enhancing machine learning methods regarding these challenges and requirements. Applications range from, but are not limited to, predictive maintenance and analytics, quality management, product design, assistance systems, optimization, and computer vision.

Topics under this session include (but not limited to)
  • Feature selection and extraction
  • Information fusion systems
  • Machine learning on resource limited hardware
  • Robust machine learning in light of uncertain and error-prone data
  • Explainable machine learning models
  • Cognitive computing
  • Reinforcement learning in industrial environments
  • Learning on streaming data
  • Adapting models in non-stationary environments
Previous Next