Mohanad Odema

I'm a Ph.D. candidate at the University of California Irvine. My research interests lie in areas relating to machine learning, efficiency, and HW/SW co-design for AI models and systems. I have hands-on expertise working on optimizations across multiple layers of abstraction, from the machine learning algorithmic level down to the underlying computing architecture.
Keywords from my projects include: Edge-AI, model compression, representation learning, generative AI, computer vision, autonomous vehicles, knowledge distillation, AutoML, Neural Architecture Search, distributed computing, HW/SW co-design, SoCs, 2.5D chiplets technology, hardware acceleration, model parallelization, compiler optimizations.

Experience

Research Assistant - 2019-2024
Department of Electrical Engineering and Computer Sciences, University of California, Irvine

PhD Intern - Summer 2023
AI Architecture and Algorithm Group, MediaTek USA

Embedded Systems Engineer - 2017-2019
Centre for Emerging Learning Technologies, The British University in Egypt

Assistant Lecturer - 2017-2019
Ain Shams University

Education

Ph.D., Computer Engineering (2019-2024)
University of California, Irvine

MSc, Computer Engineering (2014-2018)
Ain Shams University, Egypt

B.Sc., Electrical Engineering (2009-2014)
Ain Shams University, Egypt

Publications   [Scholar]

Conference Proceedings

1. SCAR: Scheduling Multi-Model AI Workloads on Heterogeneous Multi-Chiplet Module Accelerators
   Mohanad Odema, Luke Chen, Hyoukjun Kwon, Mohammad Abdullah Al Faruque
   Preprint (Under-review) (Arxiv'24)
2. Map-and-Conquer: Energy-Efficient Mapping of Dynamic Neural Nets onto Heterogeneous MPSoCs
   Halima Bouzidi, Mohanad Odema, Hamza Ouarnoughi, Smail Niar, Mohammad Abdullah Al Faruque
   2023 60th ACM/IEEE Design Automation Conference (DAC) (DAC'23)
3. SEO: Safety-Aware Energy Optimization Framework for Multi-Sensor Neural Controllers at the Edge
   Mohanad Odema, James Ferlez, Yasser Shoukry, Mohammad Abdullah Al Faruque
   2023 60th ACM/IEEE Design Automation Conference (DAC) (DAC'23)
4. EnergyShield: Provably-Safe Offloading of Neural Network Controllers for Energy Efficiency
   Mohanad Odema, James Ferlez, Goli Vaisi, Yasser Shoukry, Mohammad Abdullah Al Faruque
   Proceedings of the ACM/IEEE 14th International Conference on Cyber-Physical Systems (with CPS-IoT Week 2023) (ICCPS'23)
5. HADAS: Hardware-Aware Dynamic Neural Architecture Search for Edge Performance Scaling
   Halima Bouzidi, Mohanad Odema, Hamza Ouarnoughi, Smail Niar, Mohammad Abdullah Al Faruque
   2023 Design, Automation & Test in Europe Conference & Exhibition (DATE) (DATE'23)
6. Romanus: Robust Task Offloading in Modular Multi-Sensor Autonomous Driving Systems
   Luke Chen, Mohanad Odema, Mohammad Abdullah Al Faruque
   Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design (ICCAD'22)
7. Template Matching Based Early Exit CNN for Energy-efficient Myocardial Infarction Detection on Low-power Wearable Devices
   Nafiul Rashid, Berken Utku Demirel, Mohanad Odema, Mohammad Abdullah Al Faruque
   Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies(IMWUT'22)
8. LENS: Layer Distribution Enabled Neural Architecture Search in Edge-Cloud Hierarchies
   Mohanad Odema, Nafiul Rashid, Berken Utku Demirel, Mohammad Abdullah Al Faruque
   2021 58th ACM/IEEE Design Automation Conference (DAC)(DAC' 21)
9. EExNAS: Early-Exit Neural Architecture Search Solutions for Low-Power Wearable Devices
   Mohanad Odema, Nafiul Rashid, Mohammad Abdullah Al Faruque
   2021 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)(ISLPED' 21)
10. Energy-Aware Design Methodology for Myocardial Infarction Detection on Low-Power Wearable Devices
    Mohanad Odema, Nafiul Rashid, Mohammad Abdullah Al Faruque
    2021 IEEE/ACM International Symposium on Low Power Electronics and Design (ISLPED)(ASP-DAC' 21)

Journal Papers

1. Magnas: A mapping-aware graph neural architecture search framework for heterogeneous mpsoc deployment
   Mohanad Odema, H Bouzidi, H Ouarnoughi, S Niar, MA Al Faruque
   ACM Transactions on Embedded Computing Systems 22 (5s), 1-26 (ACM TECS '23)
2. PrivyNAS: Privacy-Aware Neural Architecture Search for Split Computing in Edge–Cloud Systems  Mohanad Odema, MA Al Faruque
   2023 Internet of Things Journal (IoTJ '23)
3. Testudo: Collaborative Intelligence for Latency-Critical Autonomous Systems  Mohanad Odema, Luke Chen, Marco Levorato, Mohammad Abdullah Al Faruque
   2022 IEEE Transactions on Computer Aided Design of Integrated Circuits and Systems (TCAD '22)
4. SAGE: A Split-Architecture Methodology for Efficient End-to-End Autonomous Vehicle Control  Arnav Malawade, Mohanad Odema, Sebastien Lajeunesse-DeGroot, Mohammad Abdullah Al Faruque
   ACM Transactions on Embedded Computing Systems (TECS) (TECS '21)