Dilsad Er

I am a Doctoral Researcher at the Max Planck Institute for Intelligent Systems. My research focuses on communication-efficient distributed learning and system-theoretic approaches to algorithm analysis. I have experience in control theory, optimization, and machine learning.

I actively contribute to outreach initiatives, including Tübingen Women in Machine Learning and Soapbox Science, promoting diversity and public engagement in STEM.

For more details, visit my Google Scholar or LinkedIn.

news

Apr 01, 2026	Our work "Teaching Machine Learning Fundamentals with LEGO Robotics" was presented by Viacheslav Sydora at the 17th International Conference on Robotics in Education (RiE) 2026!
Mar 14, 2026	We have organized the 4th TWiML Workshop, bringing together an engaged community of learners and researchers for another inspiring session!
Jan 28, 2026	Our new preprint "Teaching Machine Learning Fundamentals with LEGO Robotics" is now available on arXiv!
Dec 19, 2025	Our new preprint "A Systems-Theoretic View on the Convergence of Algorithms under Disturbances" is now available on arXiv!
Dec 01, 2025	I’ll be in San Diego for NeurIPS 2025. Feel free to stop by our workshop "DynaFront: Dynamics at the Frontiers of Optimization, Sampling, and Games". Here is my paper in the publications section.

selected publications

In prep.

Local Gradient Projection for Constrained Optimization in Deep Learning

G. D. Er, and M. Mühlebach

In preparation, 2026

Bib

arXiv

A Systems-Theoretic View on the Convergence of Algorithms under Disturbances

G. D. Er, S. Trimpe, and M. Mühlebach

arXiv preprint, 2025

Bib

@article{er2025systems,
  title = {A Systems-Theoretic View on the Convergence of Algorithms under Disturbances},
  author = {Er, G. D. and Trimpe, S. and Mühlebach, M.},
  year = {2025},
  journal = {arXiv preprint},
  url = {https://arxiv.org/abs/2512.17598}
}

ICML

Distributed Event-Based Learning via ADMM

G. D. Er, S. Trimpe, and M. Mühlebach

42nd International Conference on Machine Learning (ICML), 2025

Bib Video Poster

@article{er2024admm,
  title = {Distributed Event-Based Learning via ADMM},
  author = {Er, G. D. and Trimpe, S. and Mühlebach, M.},
  year = {2025},
  journal = {42{nd} International Conference on Machine Learning (ICML)},
  url = {https://proceedings.mlr.press/v267/er25a.html},
}