Mehmet Arif Demirtaş

I’m Arif, I am a PhD student in the Computer Science department at University of Illinois Urbana-Champaign with a focus on computing education research and human computer interaction.

I’m interested in having a better understanding of

how students learn programming
what obstacles they face and can we address these to increase participation in CS
which technologies can be utilized (e.g. AI) to aid their learning process

Currently, I’m working with Katie Cunningham to analyze how novices learn programming in non-major programming courses.

Previously, I worked on;

segmentation and localization algorithms for breast cancer detection from digital mammography imaging at Vitamu.
information extraction from loosely-structured documents using multimodal embeddings, with the NLP team at Yapi Kredi Teknoloji.
automatically lip-syncing the video of a speaker to a translated version of their speech using GANs, as part of my graduation project with Dr. Hazim Kemal Ekenel at SiMiT Lab of ITU.
end-to-end testing of autonomous driving software at UIUC with Dr. Reyhaneh Jabbarvand.
IQ score prediction from brain connectomes using graph neural networks with Dr. Islem Rekik at BASIRA Lab of ITU..
commitment concept in game theory at Cryptography, Security, and Privacy Research Group of Koc University supervised by Dr. Alptekin Kupcu.

Apart from these, I have taken part in multiple programs organized by the inzva community, both as a guide and a participant, exploring exciting topics such as generative modeling and neural radience fields and helping students from diverse areas to learn about deep learning.

Publications

Predicting cognitive scores with graph neural networks through sample selection learning

M. Hanik, M. A. Demirtaş, M. A. Gharsallaoui, and I. Rekik, “Predicting cognitive scores with graph neural networks through sample selection learning,” Brain Imaging and Behavior, vol. 16, no. 3, pp. 1123–1138, Jun. 2022, doi: 10.1007/s11682-021-00585-7.