Computer Scientist, specializing in AI, machine learning, and data science. I am passionate about developing innovative projects and continuously expanding my expertise all AI fields without limiting myself. Never stop learning!
Hello, I'm Ahmed Radwan, a Master's student and Research Assistant in Computer Science at York University. My research focuses on Distributed Machine Learning, Domain Adaptation, Real-time Optimization and AI Fairness. Previously, I was a Research Intern at KAUST, working on cutting-edge AI solutions. Passionate about AI, AR, and their real-world applications, I have over two years of experience implementing AI concepts in practical projects.
As a Research Assistant under Prof. Hina Tabassum at the LE-NGWN Lab, I am working on self-supervised learning techniques for Wi-Fi sensing and developing diffusion models for RF coverage map reconstruction. This involves designing and implementing machine learning algorithms to enhance wireless sensing capabilities and accurately reconstruct RF coverage maps, thereby improving network performance and reliability.
As a Research Intern under Prof. Tareq Y. Al-Naffouri in the Information Science Lab, I developed a real-time feedback algorithm for tracking Rak'ah completion during Salah using smartphone IMU sensors. I processed and classified sensor data for motion recognition, ensuring accurate interpretation of user movements. Additionally, I deployed and enhanced an Android app to provide real-time accuracy and error detection, improving the overall reliability and user experience of the application.
As a Research Assistant under Prof. Mohamed-Slim Alouini in the Communication Theory Lab, I optimized energy efficiency in AI tasks by employing quantization techniques to reduce model complexity and data transmission. Additionally, I analyzed centralized, federated, and split learning methods for NLP sentiment analysis, focusing on performance, privacy, and scalability. Furthermore, I led research efforts that merged AI with wireless communications, testing models on real-world noisy communication channels.
I designed and implemented LLM-based applications with a focus on Arabic language processing, conducting a comprehensive review and analysis of Arabic instruction-tuning datasets to enhance model performance.
I led a team of 50 members, guiding their development and preparing them for hackathons participation. In addition to managing the team, I conducted bootcamps on Introduction to AI, Computer Vision, and TinyML, equipping participants with essential skills. I also spearheaded the AI team for DRAG SUAS 2024, where we achieved an impressive 16th-place ranking.
I gained experience in deep learning, including autoencoders, VAEs, and GANs with a focus on unsupervised and generative modeling. I also learned reinforcement learning for optimizing policies in dynamic environments, graph neural networks for recommendation systems, and natural language processing (NLP) techniques such as text and sentiment analysis and language modeling.
I assisted in teaching the Introduction to Artificial Intelligence course, covering topics such as calculus, linear algebra, machine learning, and deep learning. Additionally, I supported the Advanced Artificial Intelligence course, which focused on CNNs, data loading, autoencoders, segmentation, and object detection.
Ahmed Y. Radwan, Hina Tabassum, Fahad Syed Muhammad
First author of a research paper enhancing CSI feedback in 3GPP’s next release using AI-driven compression.
We propose two CPC-based architectures to tackle data aging and channel variations, achieving
superior SGCS performance on industry datasets (Nokia, Oppo, CATT). Additionally,
we explore GRU pruning for real-time deployment and discuss future directions like split learning
and adaptive online strategies.
Ahmed Y. Radwan, Mustafa Yildirim, Navid Hasanzadeh, Hina Tabassum, and Shahrokh Valaee
First author of a comprehensive survey on Wi-Fi sensing, where we explore its evolution from communication routers
to sensing devices. Our work delves into channel state information (CSI) extraction, dataset comparisons,
and the impact of mobile objects on CSI. We discuss preprocessing techniques for feature extraction,
highlight the role of machine learning (ML) and self-supervised learning (SSL) in Wi-Fi sensing,
and provide a quantitative analysis of contrastive and non-contrastive learning approaches.
The paper concludes with insights into emerging technologies and future research opportunities in this field
Ahmed Y. Radwan, Mohammad Shehab, Mohamed-Slim Alouini
First author of a study on privacy-preserving, energy-efficient NLP for edge devices. We propose
semantic split learning (SL) as a TinyML framework, outperforming FL and CL in
energy efficiency and privacy, with significantly higher reconstruction error and lower CO2 emissions.
Ahmed Y. Radwan, Khaled M. Alasmari, Omar A. Abdulbagi, Emad A. Alghamdi
First author of SARD, a drag-and-drop interface for multi-chapter story generation using LLMs.
Our study shows that while node-based visualization aids mental modeling, it adds cognitive load as
stories grow. We also find AI-generated stories lack lexical diversity, highlighting patterns
and limitations for future human-AI co-writing tools.
Ahmed Y. Radwan, Layan Zaafarani, Jetana Abudawood, Faisal AlZahrani, Fares Fourati
First author of a preprint proposing a novel approach to reduce bias in AI models with
limited data. Method combines fine-tuning, ensemble learning, and knowledge
distillation, demonstrating improved fairness on CIFAR-10 and HAM10000 datasets.
Developed a novel visual interface for AI-assisted story generation, leading design, implementation, and user studies.
Co-authored preprint proposing a novel approach to reduce bias in AI models with limited data. Method combines fine-tuning, ensemble learning, and knowledge distillation, demonstrating improved fairness on CIFAR-10 and HAM10000 datasets.
Developed a model for image classification using data loading techniques and Convolutional Neural Network using Keras.
Developed and deployed a real-time Rak'ah tracking system for Salah on Android phones and watches. The project utilizes IMU sensors for motion recognition and features an enhanced Android app that provides real-time accuracy and error detection, ensuring reliable and user-friendly performance.
I will read all emails and reply to you. Send me a message, and I'll get back to you.
Please include your Name and Email Address. You will only receive a reply to your message.
