I am a PhD candidate at the Institute of Mathematics, Technische Universität Berlin, working under the supervision of Prof. Dr. Sebastian Pokutta. I am also affiliated with the Zuse Institute Berlin, where I hold a Scientific Assistant position.

I received an MSc in Mathematics in Data Science under the supervision of Prof. Dr. Michael Wolf at the Technical University of Munich in May 2020 and a BSc in Mathematics at the University of Tirana in July 2017.

My research interests lie at the interface of deep learning and optimal control, with a primary focus on developing efficient deep learning optimization algorithms of constant memory cost. Currently, I am using the Maximum Principle to study the (continuous) dynamical system approach to deep learning.

My academic resumé can be found here.

Interests

- Theoretical Deep Learning
- Robust Machine Learning
- Explainable Artificial Intelligence

Education

PhD, Mathematics, 2024 (exp.)

Technische Universität Berlin

MSc, Mathematics in Data Science, 2020

Technical University of Munich

BSc, Mathematics, 2017

University of Tirana

We present a concise optimal control optimization approach to continuous-depth deep learning models by discussing ideas and algorithms derived from the optimality conditions of the powerful Pontryagin’s Maximum Principle. The new emerging field of constant memory cost models, however, is vulnerable to adversarial attacks. Apart from highlighting the inconsistency of neural networks theoretically, we experiment with adversarial deformations for neural ordinary differential equations on MNIST and compare our results to convolutional neural-network based architectures.

- sadiku@zib.de
- Takustraße 7, Berlin, 14195
- DM Me