Shpresim Sadiku

I am interested in the theory and applications of 2D Computer Vision and Optimization. Currently, I am using first-order optimization methods to generate sparse adversarial attacks of high visual quality.

📬 Contact

office
Room 3027 at ZIB
e-mail
homepage
shpresimsadiku.com
languages
Albanian, English, German, and Serbian

🎓 Curriculum vitae

since 2020
Researcher at ZIB
May 2020
M.Sc. in Mathematics in Data Science at TUM
Jul 2017
B.Sc. in Mathematics at UNITIR

📝 Publications and preprints

Preprints

  1. Sadiku, S., Wagner, M., and Pokutta, S. (2023). Group-wise Sparse and Explainable Adversarial Attacks. [arXiv]
    [BibTeX]
    @misc{groupadversarialattack2023,
      archiveprefix = {arXiv},
      eprint = {2311.17434},
      primaryclass = {cs.CV},
      year = {2023},
      author = {Sadiku, Shpresim and Wagner, Moritz and Pokutta, Sebastian},
      title = {Group-wise Sparse and Explainable Adversarial Attacks}
    }

💬 Talks and posters

Conference and workshop talks

Sep 2022
Wavelet-based Low Frequency Adversarial Attacks
3rd BMS-BGSMath Junior Meeting, Barcelona [PDF]
Mar 2021
Neural Network Approximation Theory
BMS Conference [PDF]

Research seminar talks

Feb 2023
What Is Backpropagation?
"What Is ...?" Seminar, Berlin [PDF]

Poster presentations

Mar 2023
Wavelet-based Low Frequency Adversarial Attacks
Workshop on Optimization and Machine Learning, Waischenfeld

👨‍🏫Teaching

summer 2019
Tutor for Computational Statistics at TUM

📝 Organization and outreach

2022
Assisted in the organization of the IOL-DISCOGA Seminar in Berlin