Events

Workshop: MLaftermath

On March 9 and 10, 2026, the workshop "MLaftermath" will take place at the ECDF. ECDF Professors Emmanuel Baccelli, Professor for Open and Secure IoT Ecosystems at FU Berlin and Guillermo Gallego, Professor for Robotic Interactive Perception at TU Berlin, are co-organizers. Machine learning has become ubiquitous, as artificial neural networks (large or small) are now being deployed on almost all types of machines. The MLaftermath workshop brings together speakers from research institutions in Berlin and the surrounding region, who will present and discuss their work across the field, ranging from applied mathematics to embedded computing, experiments and datasets. Each speaker will partly base their talk on work recently published at a top AI conference (NeurIPS), with a view to fostering high-quality new scientific collaborations.

Program:

Day 1:

13:00 – Welcome

13:30 – Session 1

  • Learning Depth with Event Cameras (DERD-Net)
    Guillermo Gallego, ECDF & Technische Universität Berlin.
  • Gaussian Platting GANs
    Florian Barthel, Humboldt-Universität zu Berlin.
  • Polysemanticity
    Laura Kopf, Technische Universität Berlin.

15:00 – Coffee Break

15:30 – Session 2

  • Fusion-based RAM Optimizations for TinyML
    Zhaolan Huang, Freie Universität Berlin & Inria Berlin.
  • Additive Models
    Benedict Clark, National Metrology Institute (PTB).
  • Rethinking Explanation Evaluation under the Retraining Scheme
    Gerhard Wunder, Freie Universität Berlin.

17:00 – Networking, further chats and wrap-up 

Day 2:

10:00 – Hands-on session: TinyML tutorial (t.b.c.)

13:30 – Session 3

  • MLLM Representations
    Hanno Gottschalk, Technische Universität Berlin.
  • Efficient Quadratic Corrections for Frank-Wolfe Algorithms
    Sebastian Pokutta, Zuse-Institut Berlin.
  • Automatic Differentiation
    Tim Siebert, Humboldt-Universität zu Berlin.

15:00 – Coffee Break

15:30 – Session 4

  • tinyML Multiobjective Optimization for Urban Infrastructure Applications
    Felix Bießmann, ECDF & BHT.
  • IndEgo: A Dataset of Industrial Scenarios and Collaborative Work for Egocentric Assistants
    Vivek Chavan, Fraunhofer IPK & Technische Universität Berlin.
  • Low-Rank Training and Low-Rank Adapters
    Timon Klein, Otto-von-Guericke-Universität Magdeburg.

17:00 – Networking, further chats and wrap-up

Venue:
Einstein Center Digital Future
Conference Room, 1st Floor
Wilhelmstr. 67
10117 Berlin

Further information can be found //here.

Registration is possible //here.