PhD in EDEE @ École Polytechnique Fédérale de Lausanne, specifically at the Embedded Systems Laboratory (ESL). Previously I did my MS in Machine Learning and my undergrad in Computer Engineering and Mathematics @ AUS with a minor in Data Science. I'm Iranian and I'm 24 years old.
From May 2025 to October 2026, I was a visiting researcher at the Senseable City Lab @ MIT.
My research focuses mainly on model compression, optimization and quantization for neural networks and LLMs. I have also done studies in training optimization, computer vision, and eXplainable AI.
The picture on the right isn't what I look like now. I just wanted to put it up there so you know I've always been cool 👺
2018 – 2023: BSc. Computer Engineering, BSc. Mathematics |
American University of Sharjah
2024 – 2026: MSc. Machine Learning |
American University of Sharjah
Current: PhD in Electrical Engineering (EDEE) |
École Polytechnique Fédérale de Lausanne
I worked on the development of a multi-objective optimization framework to allow for optimization problems where a natural hierarchy exists in the "importance" of the tasks for modern deep learning architectures. You can check the project page here.
These are papers that have already been published, as reflected on my Google Scholar page (and other venues).
Transactions on Machine Learning Research (TMLR), 2026
arXiv code video paper website
We present Priority-Constrained Descent, a drop-in gradient optimizer that follows your primary objective and bends away from it only as much as needed to keep every secondary constraint progressing.
IEEE International Conference on Consumer Electronics (ICCE), 2026
We propose ER-NeRV (Error Resilient Neural Representations for Videos), a new training / finetuning framework that allows for neural representation models to be robust against hardware failures on consumer electronics.
IEEE Access, 2025
We study the class separation dynamics of Vision Transformers (ViTs) as they train for classification tasks, empirically deriving two “laws” that are always followed as a function of training. These two laws have implications that can strongly guide the training strategy and model optimization for transformer-based learning.
IEEE Communications Letters, 2025
We study the effect of quantization-aware-training (QAT) on two SOTA spectrum sensing models - DeepSense and ParallelCNN. Models are deployed on a Sony Spresense for hardware evaluation.
[IEEE] International Conference on Communication, Computing, Networking, and Control in Cyber-Phyisical Systems (CCNCPS'25), 2025
We propose a novel methodology for detecting and preventing image-scaling attacks for industrial cyber-physical systems using open data.
International Society for Photogrammetry and Remote Sensing (ISPRS), 2025
We study trends in NO2 levels through publicly available geospatial data, specifically in the Lombardy region of Italy.
IEEE Access, Vol. 11 (IEEE Engineering in Medicine and Biology Society Section), 2023
We explore eXplainable AI techniques on a Wireless Capsule Endoscopy dataset (publicly available) to better understand model decision-making and classification.
[IEEE] Advances in Science and Engineering Technology International Conferences (ASET), 2023
Designed, built and programmed a hybrid drone designed for detecting faults on the bodies of oil and gas pipelines.
[IEEE] International Conference on Communications, Signal Processing, and their Applications (ICCSPA), 2022
Studying different characteristics of two novel silicon photodiode (SiPD) devices at small- and large-scale.
[IEEE] International Seminar on Machine Learning, Optimization, and Data Science (ISMODE), 2022
Developing synthetic datasets for benchmarking the performance of feature selection algorithms. First of two works that explores the field.
These are papers that have been submitted for publication, but have not yet been released.
Transactions on Machine Learning Research (TMLR), 2026
We propose FedPCD, a priority-constrained federated optimization framework that keeps the FedAvg direction as the primary objective while applying the minimum correction necessary to meet client performance constraints, achieving the highest global accuracy across 18 heterogeneous FL benchmarks.
ACM International Conference on Multimedia (ACMMM), 2026
We investigate the content characteristics of video sequences to understand failure modes in Neural Representations for Videos.
IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2026
We propose QAT-NeRV, a quantization-aware training framework for neural video representations that integrates quantization effects directly into the training process.
Artificial Intelligence Review, 2025
We comprehensively review the quantization landscape, specifically for microcontroller-class (MCUs) devices. The survey covers advanced quantization techniques, hardware platforms as part of three families (ARM-based, RISC-V-based and NPU-based), and the software toolchains.
This includes other projects that I am currently working on, including arXiv submissions.
Unpublished, 2025
I’m in the process of writing a book on Machine Learning, to be hosted on GitHub. Compiled as part of my role as the sole TA for the Advanced Machine Learning (MLR570) graduate course offered at the American University of Sharjah.
arXiv, 2025
We study the effects of an adapted “Optimal Transport”-based quantization on Flow Matching models, showing better preservations of generating quality and latent disentanglement at extreme low-bits (sub-4 bits).
Includes presentations conducted online and in-person, and other miscellaneous academic things I've been featured in.
Khaleej Times, 2025
I was featured on an article regarding AI and its uses in academia. Feel free to take a read based on the website provided.
Google I/O Extended, 2024
Spoke at the Google I/O Extended UAE 2024 keynote, organized through Google Developer Groups Sharjah.
Teaching assistantship positions I held at the American University of Sharjah.