Selma Dilek

Selma Dilek is an assistant professor in the Department of Computer Engineering at Hacettepe University in Ankara, Turkey. She has over eight years of experience in optimization, algorithms, and the application of artificial intelligence in industrial and security systems.

Her research focuses on advanced mathematical, heuristic, and metaheuristic methods that reduce costs, increase efficiency, and enable sustainable use of resources in complex computing, embedded, and IoT systems, with strong applications in industrial processes.

• Throughout her career, Selma has worked on multiple research projects, including bilateral projects between Turkey and Thailand, as well as postdoctoral research in the UK. She has published more than 20 scientific papers in leading journals and conferences, covering topics such as:
• Optimization of scheduling and resources in complex computing and IoT systems using mathematical programming and metaheuristics.
• Efficient mapping of neural networks to Network-on-Chip (NoC) hardware accelerators for reduced communication costs and energy usage.
• Application of federated and machine learning in edge computing for IoT system security and efficiency.
• GPU-accelerated heuristic and metaheuristic algorithms (e.g., genetic algorithms, simulated annealing) for load balancing, resource optimization, and wireless sensor network coverage.
• Optimization methodologies for reliable and energy-efficient integrated chip designs.

As a lecturer in Computer Engineering, Selma teaches courses in programming, algorithms, and computer systems, actively passing her research expertise to the next generation of engineers. At the University of Bristol, her current postdoctoral work focuses on IoT system security and optimization, with an emphasis on energy-efficient and intelligent solutions for manufacturing and autonomous systems.

• Selected publications relevant to optimization:
• Simulated annealing-based high-level synthesis methodology for reliable and energy-aware ASIC designs, International Journal of Circuit Theory and Applications, 2023.
• Integer linear programming-based optimization methodology for reliability and energy-aware high-level synthesis, Microelectronics Reliability, 2022.
• A High-Level Synthesis Methodology for Energy and Reliability-Oriented Designs, IEEE Transactions on Computers, 2022.
• GPU-based parallel genetic algorithm for increasing the coverage of wireless sensor networks, IEEE ICPADS, 2017.
• Energy-aware application mapping methods for mesh-based hybrid wireless network-on-chips, Journal of Supercomputing, 2024.
• Neuron grouping and mapping methods for 2D-mesh NoC-based DNN accelerators, Journal of Parallel and Distributed Computing, 2024.
• A Novel Heuristic Neuron Grouping Algorithm for Deep Neural Network Accelerators, Journal of Circuits, Systems and Computers, 2025.
• Optimizing Deep Learning Models for Ophthalmic Disease Detection on Resource-Constrained Devices, IEEE HORA, 2025.