Dr Muhammad Umair is a Postdoctoral Researcher at the Insight SFI Research Centre for Data Analytics at the University of Galway. His current work looks at practical ways of using lightweight AI models for deepfake detection and Neural Packet Inspection (NPI), with a strong focus on real-time performance, privacy, and scalable deployment. His wider research background covers machine learning, deep learning, computational intelligence, cybersecurity, and energy analytics. Before joining Insight, he spent many years in academia as an Associate Professor of Computer Science, supervising postgraduate students and contributing to projects in applied AI, optimisation, and anomaly detection. He has published across several areas of artificial intelligence, computer vision, and intelligent systems.
Research Focus and Applications
- Developing compact deep learning models (such as MobileNetV3 and EfficientNet-Lite0) that can detect deepfakes on devices with limited computing power.
- Creating NPI approaches that combine visual analysis and network traffic patterns to verify the integrity of multimedia content in real time.
- Exploring malware detection and explainable AI, with an interest in building models that are both accurate and understandable for security analysts.
- Applying computational intelligence methods (GA, PSO, ACO, ABC) to optimisation and hybrid AI systems.
- Building efficient inference pipelines using ONNX, quantisation, and other techniques that support edge-level deployment.
- Using AI for cybersecurity and anomaly detection, as well as in energy-related applications such as energy theft detection and V2G optimisation.
- Experience across supervised learning, image processing, and intelligent system modelling, supported by a growing track record of publications in AI and smart systems.
