Hi, I’m
WAEL FATNASSI
Irvine,CA
Summary
Fourth-year PhD Candidate specializing in Autonomous Systems, ML/Deep Learning, Computer Vision, and NLP. Expertise in Reinforcement Learning, GANs, and Large Language Models. Actively seeking a full-time position for the summer of 2024 to apply my technical and educational background to real-world applications.
Overview
5
years of professional experience
1
Certification
Work History
University Of California Irvine
PhD Candidate
09.2019 - 05.2024
Job overview
Doctoral Focus: Improving the safety and reliability of autonomous vehicles and robots in uncertain environments through the practical learning algorithms and formal verification tools:
. Bernstein polynomials and neural networks’ formal verification:
Created an algorithm to efficiently approximate NN’s outputs with Bernstein polynomials, providing an upper and lower bound for the NN’s outputs. This algorithm results in more precise NN approximation ranges compared to state-of-the-art algorithms such as interval arithmetic and linear programming.
. Use of Bernstein and Neural networks to improve proposed solver:
Proposed an updated solver that uses neural networks and Bernstein polynomials to solve polynomial inequality constraints and polynomial optimization problems.
. Proposed highly parallelizable solver for polynomial inequality constraints:
Implemented a new highly parallelizable solver for polynomial inequality constraints which outperforms off-shelf solvers such as Z3 and Yices in terms of execution times.
· Created path-planning for autonomous system (Raspberry PiCar).
· Implemented a PID controller in Raspberry PiCar.
· Created ROS nodes that reads the PiCar’s position data from motion capture cameras.
University Of California Irvine
Graduate Research Assistant
05.2023 - 08.2023
Job overview
· Build a COVID-19 Detection System Using X-Rays.
· Building a Pokemon Classifier Using Transfer Learning.
· Text Generation Using Markov Chains.
· Word Embedding.
· IMDB Reviews Sentiment Analysis.
· Deciphering Text Using Character-Level RNNs.
· Emoji Predictor Using Transfer Learning in NLP.
· Build Language Model using a Recurrent Neural Network from Scratch.
University Of California, Irvine
Graduate Teaching Assistant
04.2023 - 05.2023
Job overview
- Introduced the fundamentals of perception, planning, and control of autonomous systems.
- Introduced Robot Operating System (ROS), how to integrate data from various sensors (e.g. IMU, LiDARs, and Cameras) while implementing algorithms for localization, planning, mapping, and control of autonomous vehicles to undergraduate engineering class of approx. 100 students.
- Coverd both model-based and machine learning algorithms for realizing autonomous systems
- Taught language of the ROS to students through packages and nodes while supporting their projects by building- in the Turtle Simulator ahead of time and hosting office hours to answer any questions.
- Created homework assignments for undergraduate level engineering students, balancing academic vigor with their limited technical expertise.
Education
University of California, IrvineIrvine, CA
Ph.D. from Electrical Engineering And Computer Science
05.2024
University Overview
- GPA: 4.0/4.0
- Relevant Courses: Data Structure and Algorithms, Machine Learning System Design, Deep Reinforcement Learning, GNSS Signal Processing and Software-Defined Radio Design, Network Science, Optimization Methods, Parameter and State Estimation, Learning Control Systems, Formal Methods for Autonomous Systems.
University of IdahoIdaho, USA
Master of Science from Electrical Engineering And Computer Science
05.2019
University Overview
- GPA: 3.9/4.0.
- Thesis: Learning-Based Communication Systems
- Relevant Courses: Deep Learning and Spiking Neural Networks, Fundamental Limits of Wireless Communications, Graph Theory, Machine Learning and Data Mining, Convex Optimization, Stochastic Models, Information and Coding Theory, Fourier Analysis.
Higher School of Communications (Sup’Com)Tunis, Tunisia
Bachelor of Science from Telecommunications Engineering
06.2016
University Overview
- GPA: 17.5/20.0
- Thesis Title: MIMO Systems for Increasing the Reliability of the Wireless Communications in Sensor Networks.
- Relevant Courses: Digital Communications, Signal/Image Processing, Circuits and Digital Radio Systems, Optimization, Optical Networks, Microelectronics, Hyper-Frequencies and Antennas.
Skills
- U-Net/YOLOv5/Alexnet
- ResNet/TF-IDF/Gensim
- RNN/Language Model/Tokenization
- word2vec/GloVe/TensorFlow
- Keras/CVXPY/Gurobi
- CPLEX/Autonomous Vehicles
- TensorFlow/PyTorch/z3
- Robotics/Computer Vision/Modeling
- Neural Networks /Deep Learning/Motion planning
- Imitation Learning/Reinforcement learning
- Robot Operating System (ROS)/Gazebo/SLAM
- Python/Matlab/Latex
- Keras/Scikit-learn
- Carla/MuJoCo/Vicon Tracker
- CVXPY/CUDA/Raspberry Pi
Awards and Scholarships
- 2019 Graduate & Professional Student Association (GPSA) Award, University of Idaho. One of three master’s students in the university selected that year for outstanding achievements.
- 2016 Deutscher Akademischer Austauschdienst (DAAD) Scholarship, Government of Germany. Full scholarship provided to international students to study in Germany based on their academic achievements. Awarded due to top academic performance at Sup’Com Tunisia school.
- 2014 2nd Place Winner in the International Collegiate Algorithmic Programming Contest (ACM ICPC), Sup’Com Tunisia.
- 2013 Outstanding Performance Award, IPEIT Tunisia. Ranked in the top 3% of a Tunisian national exam for math and physics in 2013. This exam is used for admission to Tunisian undergraduate engineering schools.
Certification
- Building Advanced Deep Learning and NLP Projects, Educative, Inc.
- Grokking the Machine Learning Interview, Educative, Inc.
Publications
Master Thesis:
[T] W. Fatnassi and Z. Rezki, “Learning-Based Communication Systems,” University of Idaho, pp. 73, 2019.
Journal Articles:
[J4] W. Fatnassi and Y. Shoukry, “PolyARBerNN: A Neural Network Guided Solver and Optimizer for Bounded Polynomial Inequalities,” ACM Transaction on Cyber Physical Systems, 2022, submitted.
[J3] A. Aboutaleb, W. Fatnassi, A. Chaaban, and Z. Rezki, “Optimal Diversity and Coding Gains for Millimeter-Wave Communication,” in IEEE Transactions on Vehicular Technology, vol. 70, no. 5, pp. 4601-4614, May 2021, doi: 10.1109/TVT.2021.3071330.
[J2] W. Fatnassi and Z. Rezki, “Training Deep Neural Networks for Partial Interference Cancellation in Uplink Cellular Networks,” IEEE Wireless Communications Letters, 2019, submitted.
[J1] W. Fatnassi and Z. Rezki, “Reliability Enhancement of Smart Metering System Using Millimeter Wave Technology,” in IEEE Transactions on Communications, vol. 66, no. 10, pp. 4877-4892, Oct. 2018, doi: 10.1109/TCOMM.2018.2835453.
Conference Papers:
[C9] W. Fatnassi, H. Khedr, Y. Valen, and Y. Shoukry, “BERN-NN: Tight Bound Propagation for Neural Networks using Bernstein Polynomial Interval Arithmetic,” 26th ACM International Conference on Hybrid Systems: Computation and Control (HSCC 2023).
[C8] X. Sun, W. Fatnassi, U. S. Cruz, and Y. Shoukry, “Provably Safe Model-Based Meta Reinforcement Learning: An Abstraction-Based Approach,” 2021 60th IEEE Conference on Decision and Control (CDC), 2021, pp. 2963-2968, doi: 10.1109/CDC45484.2021.9683009.
[C7] W. Fatnassi and Y. Shoukry, “PolyAR: A Highly Parallelizable Solver for Polynomial Inequality Constraints Using Convex Abstraction Refinement,” 7th IFAC Conference on Analysis and Design of Hybrid Systems (ADHS),Volume 54, Issue 5, 2021, Pages 43-48, ISSN 2405-8963, doi: 10.1016/j.ifacol.2021.08.472.
[C6] W. Fatnassi and Z. Rezki, “Upper Bound on the Expected Generalization Error of the Convolutional Neural Network,” 2020 IEEE International Conference on Communications (ICC 2020), submitted.
[C5] M.Soltani, W. Fatnassi, A.Bhuyan, Z. Rezki, and P.Titus “Physical Layer Security Analysis in The Priority-Based 5G Spectrum Sharing Systems”, 2019 Resilience Week (RWS), 2019, pp. 169-173, doi: 10.1109/RWS47064.2019.8971827.
[C4] A. Abutaleb, W. Fatnassi, M. Soltani, and Z. Rezki, “Symbol Detection and Channel Estimation Using Neural Networks in Optical Wireless Communications Systems”, CC 2019 - 2019 IEEE International Conference on Communications (ICC), 2019, pp. 1-6, doi: 10.1109/ICC.2019.8761449.
[C3] M. Soltani, W. Fatnassi, A. Abutaleb, Z. Rezki, “Autoencoder-Based Optical Wireless Communications Systems”, 2018 IEEE Globecom Workshops (GC Wkshps), 2018, pp. 1-6, doi: 10.1109/GLOCOMW.2018.8644104.
[C2] A. Aboutaleb, W. Fatnassi, A. Chaaban, and Z. Rezki, “On the Error Performance of Space-Time Codes over MIMO Nakagami Fading Channels with Blockage,” 2018 29th Biennial Symposium on Communications (BSC), 2018, pp. 1-5, doi: 10.1109/BSC.2018.8494702.
[C1] W. Fatnassi and Z. Rezki, “Increasing the Reliability of Smart Metering System Using Millimeter Wave Technology,”2018 IEEE International Conference on Communications Workshops (ICC Workshops), 2018, pp. 1-6, doi: 10.1109/ICCW.2018.8403585.
REFERENCES
Yasser Shoukry, Professor Zouheir Rezki, Professor
UC Irvine, CA UC Santa Cruz, CA
yshoukry@uci.edu zrezki@ucsc.edu
Timeline
Graduate Research Assistant
University Of California Irvine
05.2023 - 08.2023
Graduate Teaching Assistant
University Of California, Irvine
04.2023 - 05.2023
PhD Candidate
University Of California Irvine
09.2019 - 05.2024
University of California, Irvine
Ph.D. from Electrical Engineering And Computer Science
University of Idaho
Master of Science from Electrical Engineering And Computer Science
Higher School of Communications (Sup’Com)
Bachelor of Science from Telecommunications Engineering
Similar Profiles
Gretchen UgaldeGretchen Ugalde
MFA Candidate | Teaching Assistant at University Of California, IrvineMFA Candidate | Teaching Assistant at University Of California, Irvine
ANDREW SANCHEZANDREW SANCHEZ
Animal Technician at University of California IrvineAnimal Technician at University of California Irvine
Carlo ReyesCarlo Reyes
Center Attendant at University Of California, IrvineCenter Attendant at University Of California, Irvine
Caleb ChangCaleb Chang
Research Intern at UCI x GATI Paid Research Internship – University of California, IrvineResearch Intern at UCI x GATI Paid Research Internship – University of California, Irvine
Bryant FigueroaBryant Figueroa
Mortgage Loan Officer at ResMac, IncMortgage Loan Officer at ResMac, Inc