safa messaoud
Safa Messaoud
PhD Candidate
University of Illinois at Urbana-Champaign

I am a PhD candidate in Electrical and Computer Engineering, advised by Professor Alexander G. Schwing. In my thesis, I am developing methods for scalable inference and learning for structured prediction in computer vision. Previously, I have also worked on systems and healthcare analytics.

I graduated with Bachelor and Master degrees in Electrical engineering and Information technology from the Technical University of Munich. I also obtained a Master degree in Computer engineering from Virginia Polytechnic and State University, advised by Professor Sandeep Shukla. I was a visiting scholar at UC Berkeley, where I worked on my master thesis under the supervision of Professor Alberto Sangiovanni Vincentelli and Professor Andreas Herkersdorf.



July. 2021

Apr. 2021


Paper accepted at DRL4IR for oral presentation 

Paper accepted at SIGIR




Toward More Scalable Structured Models

Safa Messaoud
Ph.D. Thesis

Max-Sliced Score Matching

Safa Messaoud, Kuan-Chieh Wang, Alexander Schwing
Under submission
safa messaoud

Can We Learn Heuristics For Graphical Model Inference Using Reinforcement Learning?

Safa Messaoud, Maghav Kumar, Alexander Schwing
Computer Vision and Pattern Recognition (CVPR) 2020 -Oral Presentation-
Woman in Computer Vision Workshop (WiCV) 2020 -Oral Presentation-
Deep Vision Workshop 2020
safa messaoud

DeepQAMVS: Query-Aware Hierarchical Pointer Networks for Multi-Video Summarization

Safa Messaoud, Issmini Lourentzou, Assma Boughoula, Mona Zehni, ChangXiang Zhai , Zhizhen Jane Zhao, Alexander Schwing
Special Interest Group on Information Retrieval (SIGIR) 2021
DRL4IR Workshop (SIGIR) 2021 -Oral Presentation-
safa messaoud

Structural Consistency and Controllability for Diverse Colorization

Safa Messaoud, David Forsyth, Alexander Schwing.
European Conference for Computer Vision (ECCV) 2018
WiML Workshop at Neural Information Processing Systems (NeurIPS) 2017
safa messaoud

Medical record problem list generation

Murthy V. Devarakonda, Safa Messaoud, Ching-Huei Tsou
U.S. Patent
safa messaoud

Unsupervised analysis of transcriptomic data for demystifying the cause of Alzheimer disease

Yogatheesan Varatharajah, Mariet Younkin, Xue Wang, Arjun Athreya, Safa Messaoud, Ravi Iyer, Nilufer Ertekin-Taner
Alzheimer's Association International Conference (AAIC) 2017
safa messaoud

Reaction–diffusion modelling for microphysiometry on cellular specimens

Daniel  Grundl, Xiaorui Zhang, Safa Messaoud, Cornelia Pfister, Franz Demmel, Mario S. Mommer, Bernhard Wolf, and Martin Brischwein
Journal of Medical & Biological Engineering & Computing, 2013
Book chapter: Reaktions-Diffusions-Modelle für die Ex-vivo-Mikrophysiometrie, 2012
Bachelor Thesis (TUM)
safa messaoud

Accelerating genomic data parsing on field programmable gate arrays

Safa Messaoud, Takeshi Ogasawara
safa messaoud

Translating Discrete Time SIMULINK to SIGNAL

Safa Messaoud, Neda Saeedloei, Sandeep Shukla
Electronic Journal of Information Technology (e-TI) 2015
Talk at Arab Woman in Computing Conference (ArabWIC) 2015
Master Thesis (Virginia Tech)
safa messaoud

Optimal Architecture Synthesis for Aircraft Electrical Power Systems

Safa Messaoud
Master thesis (TUM - UC Berkeley)


Pattern Recognition - UIUC ECE 544
Machine Learning - UIUC CS 446
Mentoring - on senior thesis

Shreya Jagarlamudi (in), Alan Q. Wang (in)



2014 - 2021
GPA: 3.8/4.0
University of Illinois at Urbana-Champaign, USA
Doctor of Philosophy in Electrical and Computer Engineering
2013 - 2014
GPA: 4.0/4.0
Virginia Polytechnic and State University, USA
Master of Science in Computer Engineering
2012 - 2013
University of California Berkeley, USA
Visiting Research Scholar
2011 - 2013
GPA: 4.0/4.0
German 1.5
Technical University of Munich, GERMANY
Master of Science in Electrical Engineering and Information Technology
2008 - 2011
GPA: 4.0/4.0
German 1.3
Technical University of Munich, GERMANY
Bachelor of Science in Electrical Engineering and Information Technology


PhD Courses

Systems Courses

Programming Skills

SW: Python, MATLAB, Pytorch, Tensorflow, C/C++

HW: Verilog, VHDL, OpenCL

Work Experience


I am very curious, enjoy learning new skills and exploring new ideas, art, cultures and food. I speak four languages fluently: Arabic, French, German (FSP 1.0) and English. As a teenager, I did a lot of ceramics, silk, glass and oil painting. Later on, I found joy in computer vision research. I also play piano and enjoy reading and traveling.