Hello! Iām Lotfi, a Research Engineer at New York University Abu Dhabi, where I work with Prof. Tuka Elhanai. My role involves applying deep learning techniques to model and understand quantum systems. My research interests span Neural Architecture Search (NAS), which is about teaching AI to improve itself, deep learning principles, the emerging field of quantum computing, and computer vision, which enables machines to interpret visual information as humans do.
My long-term research objective is the creation of the `Master Algorithm`: a universal AI framework capable of tackling a broad spectrum of challenges. I believe that NAS, which automates the design of highly efficient and adaptable AI models, is fundamental to this pursuit. In the short term, my focus is on developing NAS methods that excel in various domains and exploring ways to transfer this knowledge seamlessly across different fields.
October 2023 - Now
Domains: AutoML, Quantum Computing, Deep Learning
October 2022 - September 2023
Domains: Efficient Deep Learning, Edge Computing
Role: Lead Author
Role: Project Lead
Role: Second Author
September 2021 - September 2022
Domains: Neural rchitecture Search, Computer Vision, Edge Computing, Transformers
2020 - 2021
Mecharbat Lotfi Abdelkrim, Hadjer Benmeziane, Kaouther El Maghraoui ,Hamza Ouarnoughi, Smail Niar
we introduce DiffNAS, a novel NAS methodology rooted in diffusion processes. DiffNAS brings substantial improvements in both NAS search efficiency and the quality of the generated neural network architectures. To the best of our knowledge, our work marks a pioneering effort in applying diffusion algorithms to enhance the search space exploration with NAS. Our experimental results, conducted on the widely-used NAS-Bench-101, showcase the remarkable capabili- ties of DiffNAS. We achieved the highest average accuracy, outperforming other state-of-the-art methods, while completing the search process at least 2 times faster.
Paper PresentationMecharbat Lotfi Abdelkrim, Hadjer Benmeziane, Smail Niar, Hamza Ouarnoughi
We developed HyT-NAS, an efficient Hardware-aware Neural Architecture Search (HW-NAS) including hybrid architectures targeting vision tasks on tiny devices. HyT-NAS improves state-of-the-art HW-NAS by enriching the search space, enhancing the search strategy as well as the performances predictors.
Paper PresentationI am a Huawei Certified ICT Associate in Artificial Intelligence after completing the exam that covers the different techniques of artificial intelligence, especially machine learning, artificial neural networks and Deep Learning.
I completed Datacamp's datascience track that contains more than 20 courses that cover all theoretical and practical aspects in machine learning and its application on different domains such as Computer vision and Natural Language processing.