Mohamad Arafeh
Research Subjects
Mohamad Arafeh has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Azzam Mourad
Khalifa University
4 shared publications - Hadi Otrok
Khalifa University
4 shared publications - Hakima Ould-Slimane
Department of Mathematics and Computer Science, Universit du Qubec Trois-Rivires, Trois-Rivires, QC, Canada
3 shared publications - Hani Sami
Concordia Institute for Information Systems Engineering, Concordia University, Montreal, QC, Canada
3 shared publications - Mohamad Wazzeh
Department of Software and IT Engineering, cole de technologie suprieure (TS), Montreal, QC, Canada
3 shared publications - Ahmad Hammoud
Department of CSM, Artificial Intelligence and Cyber Systems Research Center, Lebanese American University, Beirut, Lebanon
2 shared publications
Latest Publications
- 18 January 2025
- Published by Elsevier in Journal of Network and Computer Applications
- Vol. 236
Efficient privacy-preserving ML for IoT: Cluster-based split federated learning scheme for non-IID data
- 18 January 2025
- Published by Elsevier in Journal of Network and Computer Applications
- Vol. 236
Efficient privacy-preserving ML for IoT: Cluster-based split federated learning scheme for non-IID data
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (2) , 1974-1989
WFSL: Warmup-Based Federated Sequential Learning
Federated learning (FL) gained importance in sensitive Internet of Things (IoT) environments by creating a privacy-preserving ecosystem where participants share machine-learning models instead of raw data. However, FL shifts data control away from the server, exposing it to non-independent and identically distributed (non-IID) problems caused by biased clients (IoT devices). This hinders the learning process by increasing execution time and cost. Current solutions alter the FL structure or compromise privacy by offloading clients’ raw data to an external server. To mitigate these limitations, this article proposes a solution to the non-IID problem by introducing an initialization phase, orchestrated by the server, that constructs high-quality initial models. These models can boost FL accuracy and convergence, regardless of whether IoT participants exhibit non-IID properties. Our proposed initialization scheme involves clients training over the same model sequentially, lessening the impact of aggregation, a primary cause of model degradation in federated approaches. Furthermore, a regulator algorithm deployed on the server maintains model integrity and mitigates catastrophic forgetting, enhanced by a client selection process that emphasizes the compatibility of IoT clients to cooperate effectively. Moreover, we devise an optimization scheme based on clustering and genetic algorithms to reduce the selection time while ensuring optimal performance in IoT networks. Experiments on MNIST, KDD, and CIFAR10 data sets show promising results in terms of initial model resiliency against catastrophic forgetting and non-IID settings. Additionally, our findings suggest that our approach can significantly enhance FL training in IoT applications by achieving 40% higher initialization accuracy and a 20% average improvement in end results compared to conventional methods, all while reducing computation time by 80% compared to similar approaches.
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (2) , 1974-1989
WFSL: Warmup-Based Federated Sequential Learning
Federated learning (FL) gained importance in sensitive Internet of Things (IoT) environments by creating a privacy-preserving ecosystem where participants share machine-learning models instead of raw data. However, FL shifts data control away from the server, exposing it to non-independent and identically distributed (non-IID) problems caused by biased clients (IoT devices). This hinders the learning process by increasing execution time and cost. Current solutions alter the FL structure or compromise privacy by offloading clients’ raw data to an external server. To mitigate these limitations, this article proposes a solution to the non-IID problem by introducing an initialization phase, orchestrated by the server, that constructs high-quality initial models. These models can boost FL accuracy and convergence, regardless of whether IoT participants exhibit non-IID properties. Our proposed initialization scheme involves clients training over the same model sequentially, lessening the impact of aggregation, a primary cause of model degradation in federated approaches. Furthermore, a regulator algorithm deployed on the server maintains model integrity and mitigates catastrophic forgetting, enhanced by a client selection process that emphasizes the compatibility of IoT clients to cooperate effectively. Moreover, we devise an optimization scheme based on clustering and genetic algorithms to reduce the selection time while ensuring optimal performance in IoT networks. Experiments on MNIST, KDD, and CIFAR10 data sets show promising results in terms of initial model resiliency against catastrophic forgetting and non-IID settings. Additionally, our findings suggest that our approach can significantly enhance FL training in IoT applications by achieving 40% higher initialization accuracy and a 20% average improvement in end results compared to conventional methods, all while reducing computation time by 80% compared to similar approaches.
- 3 August 2024
- Published by Elsevier in Journal of Network and Computer Applications
- Vol. 231
CRSFL: Cluster-based Resource-aware Split Federated Learning for Continuous Authentication
- 3 August 2024
- Published by Elsevier in Journal of Network and Computer Applications
- Vol. 231
CRSFL: Cluster-based Resource-aware Split Federated Learning for Continuous Authentication
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Surveys & Tutorials
- Vol. 26 (4) , 2914-2960
The Metaverse: Survey, Trends, Novel Pipeline Ecosystem & Future Directions
The Metaverse offers a second world beyond reality, where boundaries are non-existent, and possibilities are endless through engagement and immersive experiences using the virtual reality (VR) technology. Many disciplines can benefit from the advancement of the Metaverse when accurately developed, including the fields of technology, gaming, education, art & culture, socialization, commerce, and businesses. Nevertheless, developing the Metaverse environment to its full potential is an ambiguous task that needs proper guidance and directions. Existing surveys on the Metaverse focus only on a specific aspect and discipline of the Metaverse and lack a holistic view of the entire process. Moreover, most surveys refrain from providing detailed guidance about the development process of the metaverse, including its impact on technologies, businesses, existing challenges, and potential research directions due to their lack of a macro and micro perception of such a topic. To this end, a more holistic, multi-disciplinary, in-depth, and academic and industry-oriented review is required to provide a thorough study of the Metaverse development pipeline and fill the gap in existing Metaverse surveys. To address these issues, we present in this survey a novel multi-layered pipeline ecosystem composed of (1) the Metaverse computing, networking, communications and hardware infrastructure, (2) environment digitization, and (3) user interactions. For every layer, we discuss the components that detail the steps of its development. Also, for each of these components, we examine the impact of a set of enabling technologies and empowering domains (e.g., Artificial Intelligence, Security & Privacy, Blockchain, Business, Ethics, and Social) on its advancement. In addition, we explain the importance of these technologies to support decentralization, interoperability, user experiences, interactions, and monetization. Our presented study highlights the existing challenges for each component, followed by research directions and potential solutions. To the best of our knowledge, this survey is the most comprehensive and allows users, scholars, and entrepreneurs to get an in-depth understanding of the Metaverse ecosystem to find their opportunities and potentials for contribution.
- 23 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Communications Surveys & Tutorials
- Vol. 26 (4) , 2914-2960
The Metaverse: Survey, Trends, Novel Pipeline Ecosystem & Future Directions
The Metaverse offers a second world beyond reality, where boundaries are non-existent, and possibilities are endless through engagement and immersive experiences using the virtual reality (VR) technology. Many disciplines can benefit from the advancement of the Metaverse when accurately developed, including the fields of technology, gaming, education, art & culture, socialization, commerce, and businesses. Nevertheless, developing the Metaverse environment to its full potential is an ambiguous task that needs proper guidance and directions. Existing surveys on the Metaverse focus only on a specific aspect and discipline of the Metaverse and lack a holistic view of the entire process. Moreover, most surveys refrain from providing detailed guidance about the development process of the metaverse, including its impact on technologies, businesses, existing challenges, and potential research directions due to their lack of a macro and micro perception of such a topic. To this end, a more holistic, multi-disciplinary, in-depth, and academic and industry-oriented review is required to provide a thorough study of the Metaverse development pipeline and fill the gap in existing Metaverse surveys. To address these issues, we present in this survey a novel multi-layered pipeline ecosystem composed of (1) the Metaverse computing, networking, communications and hardware infrastructure, (2) environment digitization, and (3) user interactions. For every layer, we discuss the components that detail the steps of its development. Also, for each of these components, we examine the impact of a set of enabling technologies and empowering domains (e.g., Artificial Intelligence, Security & Privacy, Blockchain, Business, Ethics, and Social) on its advancement. In addition, we explain the importance of these technologies to support decentralization, interoperability, user experiences, interactions, and monetization. Our presented study highlights the existing challenges for each component, followed by research directions and potential solutions. To the best of our knowledge, this survey is the most comprehensive and allows users, scholars, and entrepreneurs to get an in-depth understanding of the Metaverse ecosystem to find their opportunities and potentials for contribution.