Data forms the basis of many modern machine learning approaches, but what happens when data is either incomplete, incomplete or not available at all? This issue is becoming increasingly relevant in data-poor contexts such as rare medical diagnoses, highly specialized industrial applications or the modelling of unknown systems. The presentation will highlight current scientific strategies and methods that enable learning without or with minimal data. The focus is on approaches such as zero-shot and few-shot learning, domain-specific modeling, transfer learning and the use of synthetic data and generative models. In addition, the theoretical and practical challenges associated with data-poor scenarios are discussed.
With the METECH Lecture Series, we offer scientists from Chemnitz University of Technology, Ilmenau University of Technology and OVGU Magdeburg a platform for interdisciplinary exchange and networking. The METECH Lecture Series focuses on current research topics
from the field of people and technology. Our lectures consist of a 30-minute presentation + 30 minutes for questions and discussion. All presentations will be held in a hybrid format so that everyone can participate both in person and virtually.
We look forward to seeing you!
