German Magai

Email: gera128a@gmail.com
Website: https://github.com/Germandev55
Link to CV: https://drive.google.com/file/d/12rd8N2p1QMvIR7jPkMXnoXclbzGMgUgt/view?usp=sharing
Level: PhD, Internship, Research Visit
Keywords: Deep learning theory, Computer vision (CV), Topology and geometry in deep learning, topological data analysis (TDA), Generative models, Face recognition and object detection
Short bio / research interests / publications: I am a second year Ph.D. student at the Higher School of Economics (Moscow), and I am interested in deep learning, computer vision, and algebraic topology methods in computer science. Previously, I worked as a research engineer in the field of computer vision. The topic of my master’s thesis was related to face recognition systems and methods of protection against attacks on these systems. I am looking for a PhD position or a research engineer position.

My research interests include the following topics:
1) Theoretical issues of deep learning (interpretability and explainability of deep neural network (DNN) models, convergence, evaluation of generalizing ability)
2) Application of algebraic topology and geometry methods to analyze deep learning models (CNNs, Vision Transformers) and evaluate the quality and performance of DNN models.
3) Generative learning. GAN performance evaluation problems and other related theoretical issues.
4) Face recognition systems, anti-spoofing algorithms, as well as object recognition and detection (gestures, clothes, etc.)