Lecture Series “Robust AI”: Sascha Saralajew

On Thursday, April 25, we will continue our lecture series “Robust AI”. Dr. Sascha Saralajew will give a lecture on “Prototype-based Classification Learning: From traditional versions to modern approaches”.

When & where

25 April 2024, 4pm, online: https://uni-bielefeld.zoom-x.de/j/64775735478?pwd=TFpEUVFPME5EQXFKMHZHY1ZsM2Y4Zz09

Abstract

Prototype-based Classification Learning is a powerful machine learning method with the great properties of being interpretable and provably robust at the same time. These properties make the method an interesting approach for several domains (e.g., biomedical) and, particularly, for domains that are considered to be high-risk in the EU AI Act. In this talk, I will introduce the concept of prototype-based classification learning, starting from the traditional methods (e.g., learning vector quantization) to modern approaches that fuse prototype-based classification concepts with deep neural networks (e.g., ProtoPNet and prototype-based large language models). Throughout this introduction, I will focus on the properties of prototype-based classification methods, such as interpretability, and whether they are preserved in modern approaches and other variations. I will give examples, show opportunities, and will talk about open questions.

Speaker

Sascha Saralajew is a Senior Research Scientist at NEC Laboratories Europe in Heidelberg (https://scholar.google.com/citations?user=YTi93_0AAAAJ)