Dr. Fares Meghdouri

Deep Learning, Computer Vision, Network Security, Astronomy and Electronics.

prof_pic.jpg

Vienna, Austria

Hello there, I am Fares, and welcome to my small space. I am a deep learning researcher and computer vision software engineer. I also do astrophotography in my spare time, build small embedded electronic projects, and read. I’ve been a researcher at the Technical University of Vienna (TU Wien), where I obtained my PhD and mainly focused on applied machine learning for cyber security and anomaly detection in communication networks.

I share my thoughts, publications, hobby projects, and more on this website. Be aware that I only share my own thoughts and opinion, so don’t be offended if you do not share the same opinion. Enjoy!

Wish to discuss/share ideas? contact me at meghfares@hotmail.com.


news

Feb 23, 2024 You have a PDF and want to know if it contain sensitive infomation and maybe want to blend it? Checkout my latest implementation of such a system on my Github under NER. It also support scanned images!
Jan 1, 2024 New year and a new idea. Checkout the ImageLookup repository on my Github. It’s an implementation of an idea that performs fast image lookup based on similarity. Imagine you have an image or subimage and have a huge database of reference images, use this to find the best match quickly.
Nov 26, 2023 Well, I’ve been busy with work for a year… I am glad to announce that I’ve finally obtained my PhD and currently focusing more on machine learning (which is now called AI - a thing that I don’t like but yeah, whatever!). I’ve now added some reporsitories and ideas that I’ve been working on in my free time recently.
Dec 18, 2022 I’ve now started implementing YoloV3 in C++. The goal is to get familiar with the language and imporve the FPS from my last python implementation. Currently focusing on inference. Follow my progress here.
Dec 5, 2022 I’ve started recently working on a YoloV3 implementation from scratch with pytorch. Check the progress here. Update 09.12.22: the implementation is finished and I am currently workig on imporving the FPS and implementing the training loop.

selected publications

  1. Cross-Layer Profiling of Encrypted Network Data for Anomaly Detection
    Fares MeghdouriFélix Iglesias Vázquez, and Tanja Zseby
    In 2020 IEEE 7th International Conference on Data Science and Advanced Analytics (DSAA) 2020
  2. Controllable Network Data Balancing With GANs
    Fares MeghdouriThomas SchmiedThomas Gärtner, and Tanja Zseby
    In NeurIPS 2021 Workshop on Deep Generative Models and Downstream Applications 2021