UAAT International Young Visiting Scholar Program
Muhammad Tayyab Noman
Current Position/Title: Senior Researcher/Dr.
Institutional Affiliation: Technical University of Liberec, Czech Republic
Email: muhammad.tayyab.noman@tul.cz; tayyab_noman411@yahoo.com
Webpage: https://www.researchgate.net/profile/Muhammad-Noman-38?ev=hdr_xprf
Host Scholar: Deepak Balram, Research Assistant Professor Hosting Department/Institution: Department of Electrical Engineering, National Taipei University of Technology.
Biography:
Dr. Muhammad Tayyab Noman is pursuing his career as a Senior Researcher at Technical University of Liberec, Czech Republic. Through his research, he is trying to understand the reaction dynamics of photocatalytic charge carriers; modelling of geometry and simulation of process variables. He is proficient in material analysis and simulation software, i.e., Python, ANSYS, Origin Pro, and Design Expert, and have extensive experience in mechanical testing, failure analysis, surface analysis, and machine learning (ML).
Lecture [1]:
Time: June 19, 2025, 16:00–17:30
Venue: Conference room, Department of Electrical Engineering, National Taipei University of Technology.
Title: The role of machine learning during the fabrication of photoactive functional nanocomposites
Abstract:
The fabrication of photoactive functional nanocomposites has emerged as a transformative area in materials science, offering promising applications in energy conversion, environmental remediation, and sensing technologies. However, optimizing their synthesis parameters to achieve desired structural, optical, and photocatalytic properties remains a complex challenge due to the multidimensional nature of materials processing. In this context, ML has become a powerful tool to accelerate the design and fabrication of nanocomposites by identifying patterns, predicting outcomes, and guiding experimental decisions with minimal trial-and-error. This lecture explores the integration of ML algorithms, e.g., supervised learning, regression models, neural networks, and ensemble methods, into the fabrication workflow of photoactive nanocomposites. The applications where ML methods are used include the prediction of bandgap energies, photocatalytic efficiencies, particle size distributions, and the relationship between synthesis parameters and material performance. The lecture highlights recent advancements, challenges, and future prospects of ML-assisted fabrication, ultimately underscoring its potential to revolutionize the development of next-generation photoactive materials.
Lecture [2]:
Time: August 6, 2025, 16:30–18:00
Venue: Conference room, Department of Electrical Engineering, National Taipei University of Technology.
Title: Machine learning methods used in composites
Abstract:
The integration of ML in composite materials research has revolutionized the way materials are designed, analyzed, and optimized. This lecture provides a comprehensive overview of machine learning methods employed in the field of composite materials, highlighting their applications in property prediction, failure analysis, microstructure optimization, and process parameter tuning. Supervised learning algorithms such as support vector machines, decision trees, random forests, and neural networks have shown high accuracy in predicting mechanical, thermal, and electrical properties based on compositional and processing parameters. Unsupervised learning and clustering 4 | 4 methods are used to uncover hidden patterns in large experimental datasets, while deep learning and convolutional neural networks have emerged as powerful tools for microstructural image analysis and defect detection. Furthermore, reinforcement learning and optimization techniques are being increasingly utilized for automated materials design and multi-objective optimization. This lecture underscores the potential of ML to accelerate the development of nextgeneration composite materials, reduce experimental costs, and support datadriven decision-making in materials science. Challenges related to data availability, model interpretability, and domain integration are also discussed, along with future directions for research.