Contact details
Links
Scientific classifications
- 1. Natural sciences
- 1.1 Mathematics
- Applied mathematics
- 1.7 Other natural sciences
- 1.1 Mathematics
- 1.2 Computer and information sciences
- Computer sciences
- bioinformatics
- information science
Main research areas
My research focuses on the innovative application of machine and deep learning methods. The two main directions of the study are as follows: optimal resource use during the application of neural models, i.e. reducing the amount of computing resources needed to solve tasks, improving their business application, and serving sustainability goals through lower energy consumption. The other main direction is the application of deep learning in new, hitherto untouched application areas. During all of this, research and development work touches on many topics, such as machine vision, segmentation tasks, natural language processing, and processing sequences/logs, and uses most technologies in deep learning, like convolutional, recurrent, and transformer architectures, and different training approaches.
Highlighted publications
- 2007 – Impact of non-Poissonian activity patterns on spreading processes – mtmt.hu
- 2020 – Attention U-net based adversarial architectures for chest X-ray lung segmentation – mtmt.hu
- 2024 – Deep learning the Hurst parameter of linear fractional processes and assessing its reliability – mtmt.hu
- 2025 – Estimating room acoustic descriptors from bag-of-vectors representation with transformers – mtmt.hu
- 2024 – HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing – mtmt.hu