Inductive Bias of Deep Neural Networks

My name is Lukas Neumann, I am an Assistant Orofessor in the Visual Recognition Group (CMP) of Czech Technical University in Prague (CTU). I got my PhD degree under supervision of prof. Jiří Matas and then I was a postdoctoral researcher in Visual Geometry Group (VGG), University of Oxford under supervision of prof. Andrea Vedaldi and prof. Andrew Zisserman.

I am currently looking for passionate and motivated colleagues to join my new research team at Czech Technical University (CTU) in cutting-edge research of Artificial Intelligence.

We have open PhD and postdoctoral positions for individuals driven to deepen fundamental understanding of AI — not just those happy with "it works, but we don’t know why." Our aim is to go beyond applications and investigate the core principles and theoretical foundations of AI and computer vision systems.

Research topics include (but are not limited to):

• Training-free Network Architecture Search (NAS)
• World models for computer vision (JEPA, ...)
• Energy efficiency of modern network architectures
• Algorithms for network architecture optimization
• Efficient and Compact Deep Network Designs

If you are excited by foundational questions in AI and eager to contribute to high-impact research, I’d love to hear from you! Please send your applications at lukas.neumann@cvut.cz for more information!