We Plan to Grow! Exciting Opportunities Ahead in Hybrid Artificial Intelligence

The Institute of Software Engineering and Programming Languages (ISP), University of Lübeck, is thrilled to announce an upcoming expansion in our research endeavors. Under the leadership of Dr. Gesina Schwalbe, we are planning to form a young researcher's group on Correctable Hybrid Artificial Intelligence.

About the Initiative

Funded by the German Ministry for Education and Research, our group will focus on developing innovative methods for the verification, integration, and correction of symbolic knowledge within deep neural networks. Subject to funding approval, this initiative is set to run from September 2024 to August 2027 and aims to make significant strides towards creating trustworthy artificial intelligence.

Research Areas

We are looking for individuals with a solid background in computer science or mathematics who are interested interested in research towards a doctoral degree in one of the following topic areas:

•     Verification and Integration of Symbolic Constraints: Exploring how to embed concept vectors into the latent space of deep neural networks to ensure safe and trustworthy AI.
•     Subnetworks and Logical Units in DNNs: Developing methods for the representation, insertion, and correction of logical subtasks within deep neural networks.
•     Guaranteed DNN Performance: Verifying neural network performance against symbolic knowledge bases for applications in natural language processing and robotics.

You are interested? Great!

Please contact Dr. Gesina Schwalbe (gesina.schwalbe [at] uni-luebeck.de (subject: Correctable%20Hybrid%20Artificial%20Intelligence) ) to discuss further steps, including application procedure and availability of funding.

Join us as we grow and shape the future of artificial intelligence!