Welcome!

Hi, I’m Janek Haberer, a PhD student in Computer Science at Kiel University. My work focuses on the design of efficient deep learning methods in resource-constrained environments, with research areas spanning Edge AI, dynamic vision transformers, federated learning, and progressive image compression. I have 3 years of experience with 7 published papers in peer-reviewed conferences and journals: NeurIPS, CVPR Workshop, IEEE Access, MobiSys, CoNEXT Workshop, and EWSN Workshop. I am passionate about advancing machine learning technologies and collaborating on impactful projects.

Featured Publication

HydraViT: Stacking Heads for a Scalable ViT

J. Haberer, A. Hojjat, O. Landsiedel
NeurIPS'24: Advances in Neural Information Processing Systems 37, 2024

This work introduces HydraViT, a novel approach to scaling Vision Transformers (ViTs) by stacking multiple attention heads, enabling efficient and scalable deep learning for resource-constrained environments.
Published Version | OpenReview | arXiv | GitHub

Quick Facts

• Current Role: Graduate Researcher, Kiel University
• Research Interests:
  • Edge AI
  • Dynamic Vision Transformers
  • Federated Learning
  • Progressive Image Compression
• Skills: Python, PyTorch, TensorFlow
• Awards: Best Presentation Award (EMERGE Workshop at EWSN'24)