Ege Erdogan

MSc Informatics at the Technical University of Munich

Hey! I am Ege, an informatics MSc student at TUM currently writing my thesis with David Rügamer and Bastian Rieck. I am broadly interested in geometric DL, deep generative models, and safety/interpretability of ML systems.

I was previously an undergrad (CS & Math) at Koç University in Istanbul, Turkey, where I worked on projects related to data/model confidentiality in collaborative ML as an intern at the Cryptography, Security, and Privacy Research Group.

Feel free to contact me at ege at erdogan.dev.

news

Jan 2025	🎉 Finished writing my masters thesis! Click for the PDF
Jun 2024	📕 Starting my masters thesis on efficient sampling of NN weights using generative models and geometric DL, with David Rügamer and Bastian Rieck.
Jun 2024	🎉 Paper accepted at CANS ‘24. SplitOut: Out-of-the-Box Training-Hijacking Detection in Split Learning via Outlier Detection
Oct 2023	🧪 Starting TUM-DI-LAB, working on geometric generative models for molecules.
Oct 2023	🎉 Paper accepted at the GLFrontiers Workshop within NeurIPS ‘23. Poisoning × Evasion: Symbiotic Adversarial Robustness for Graph Neural Networks

selected publications

CANS ’24

SplitOut: Out-of-the-Box Training-Hijacking Detection in Split Learning via Outlier Detection

Erdogan, Ege, Teksen, Unat, Celiktenyildiz, Mehmet Salih, Kupcu, Alptekin, and Cicek, A. Ercument

International Conference on Cryptology And Network Security (CANS ’24) 2024

Website
GLFrontiers ’23

Poisoning × Evasion: Symbiotic Adversarial Robustness for Graph Neural Networks

Erdogan, Ege, Geisler, Simon, and Günnemann, Stephan

NeurIPS 2023 Workshop: New Frontiers in Graph Learning 2023

Website
(SRW) RANLP ’23

Detecting ChatGPT: A Survey of the State of Detecting ChatGPT-Generated Text

Dhaini, Mahdi, Poelman, Wessel, and Erdogan, Ege

In Proceedings of the 8th Student Research Workshop associated with the International Conference Recent Advances in Natural Language Processing 2023

WPES ’22

SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learning

Erdogan, Ege, Kupcu, Alptekin, and Cicek, A Ercument

The 21st Workshop on Privacy in the Electronic Society 2022

Bib Website

@article{erdogan2021splitguard,
  title = {SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learning},
  author = {Erdogan, Ege and Kupcu, Alptekin and Cicek, A Ercument},
  journal = {The 21st Workshop on Privacy in the Electronic Society},
  year = {2022},
}

WPES ’22

UnSplit: Data-oblivious model inversion, model stealing, and label inference attacks against split learning

Erdogan, Ege, Kupcu, Alptekin, and Cicek, A Ercument

The 21st Workshop on Privacy in the Electronic Society 2022

Bib Website

@article{erdogan2021unsplit,
  title = {UnSplit: Data-oblivious model inversion, model stealing, and label inference attacks against split learning},
  author = {Erdogan, Ege and Kupcu, Alptekin and Cicek, A Ercument},
  journal = {The 21st Workshop on Privacy in the Electronic Society},
  year = {2022},
}

SRDS ’21

(Demo) Zelig: Customizable Blockchain Simulator

Erdogan, Ege, Aydin, Can Arda, Ozkasap, Oznur, and Gill, Waris

40th International Symposium on Reliable Distributed Systems 2021

arXiv Bib

@article{erdogan2021zelig,
  title = {(Demo) Zelig: Customizable Blockchain Simulator},
  author = {Erdogan, Ege and Aydin, Can Arda and Ozkasap, Oznur and Gill, Waris},
  journal = {40th International Symposium on Reliable Distributed Systems},
  year = {2021},
}