Murtadha Hssayeni

The main role is developing deep learning models for detecting cyberattacks on Internet of Things (IoT) networks that plays a pivotal role in advancing both academic research and practical cybersecurity applications. The task is crafting and refining deep learning models. This involves a multi-dimensional approach, encompassing the exploration of novel neural network architectures, feature extraction techniques, and optimization strategies. By meticulously fine-tuning these models, I contribute to the development of robust and efficient algorithms capable of swiftly recognizing and categorizing complex cyberattacks targeting IoT networks. This role demands a profound understanding of both the nuances of deep learning methodologies and the intricacies of IoT network vulnerabilities.

In addition to model development, I am a prolific contributor to the academic community through the creation of comprehensive and insightful scientific papers. These papers serve as a conduit for sharing knowledge, methodologies, and findings, contributing to the ever-growing body of cybersecurity research. The responsibility to clearly articulate experimental setups, results, and implications ensures the dissemination of valuable insights to peers, enabling collective progress in IoT network security.

Another significant aspect of my role is the allocation and management of honeypots. These controlled decoy systems are strategically designed to attract cyber attackers, allowing the postdoc to observe, analyze, and gain a deeper understanding of evolving attack vectors and tactics. This process aids in refining the developed deep learning models by incorporating real-world attack patterns and enhancing their ability to detect sophisticated threats.

Generative Adversarial Networks (GAN) to deal with limited time-series data in regression problems
Machine learning algorithms for in-home therapeutic management systems for patients with neurodegenerative diseases (Parkinson’s and Alzheimer’s diseases)
COVID-19 risk estimation using Recurrent Neural Networks (RNN)
Pattern discovery of cognitive response captured using functional near-infrared spectroscopy (fNIRS) signals in infants
The first author of several journal and conference publications
Prepared Python lectures for intro to deep learning course
Worked as a teaching assistant for many undergrad labs