Profile Picture

Email: kamil.erdayandimarmaraedutr
Office No: M2-232
Office Tel: +90 216 777 35 35

Google Scholar
ORCID iD icon0000-0001-9507-662X
Introduction


Kamil Erdayandı is a Faculty Member in the Computer Engineering Department at Marmara University.

He obtained his Ph.D. in Computer Science from The University of Manchester, under the supervision of Dr. Mustafa A. Mustafa. His thesis was titled "Privacy-Preserving Trading in Local Energy Markets". He then worked as a Research Associate on the EnnCore Project, within the S3 research group at The University of Manchester.

His research interests include Multiagent Systems, Data Privacy, Applied Cryptography, Energy Trading, Federated Learning and Blockchain.

Latest News (June 2025)
  • I'll be serving as a co-chair in Energy track at ICDLT'25
Potential Projects for BSc Students

I will be more than happy to work with BSs students on these project topics:

Engineering Projects for BSc Students (for those taking CSE4197/CSE4297 course units):
  • AI Agents for multi-player games: This project line focuses on designing intelligent AI agents that can play games optimally, much like the opponents we encounter when playing against a computer. The goal is to build agents capable of strategic decision-making in competitive environments. The project could incorporate advanced AI techniques such as Reinforcement Learning, Monte Carlo Tree Search (MCTS), AlphaZero, and MuZero.
  • Trading mechanism design: This project focuses on developing trading and pricing mechanisms within multi-agent environments. Students will design interaction protocols and distributed optimisation strategies, potentially using techniques from game theory to model and analyse agent behaviour. For this project, an optional extension could include incorporating blockchain technology for decentralisation.
  • Secure Federated Learning: This project focuses on detecting backdoor attacks in federated learning environments, where certain agents may attempt to manipulate the global model during collaborative training. Students will develop techniques to identify such malicious behaviour and ensure model integrity in distributed settings. Through this project, students will gain hands-on experience in data science, machine learning, and data privacy.
Teaching
Current:
  • Lecturer @ Marmara University: EE2032 Electronics, CSE4095 Special Topics in Computer Engineering III (Multi-Agent Systems)
Previous:
  • Teaching Assistant @ The University of Manchester: COMP61411 Cryptography, COMP34612 Computational Game Theory, COMP34111 AI and Games, COMP63342 Software Security, COMP61011 Foundations of Machine Learning, COMP60532 Principles of Digital Biology, COMP60542 Introduction to Health Informatics, COMP26020 Programming Languages & Paradigms, COMP26120 Algorithms and Data Structures, COMP32412 The Internet of Things, COMP24412 Knowledge Based AI
  • Teaching Assistant @ Yıldız Technical University: BLM1612 Circuit Theory, BLM2622 Electronic Circuits, BLM 2611 Logic Circuits, BLM 4920 Real-Time Computer Systems, BLM3621 Microprocessors
Publications ‑ Peer Reviewed & Published

Under Construction .../


The publications in Google Scholar page are up-to-date.


Preprints, Datasets, Tools, Misc.

Under Construction .../


Data-Generation: A tool designed to generate data for household electricity consumption and solar panel power generation. [Relevant paper, Source code]

PP-LEM and PFET: Clearance Mechanisms for Local Energy Markets. [Relevant papers: PP-LEM, PFET; Source code]