Kokulan Thangasuthan

Research project title: 
Strategic Adaptive Learning: A Game-Theoretic Approach to Navigating Uncertainty
Principal goal for project: 
This project aims to bridge the gap between theoretical game theory and practical, real-world applications, particularly in dynamic and unpredictable environments. By developing a comprehensive framework that combines the precision of machine learning and the strategic insights of game theory, my research seeks to enhance decision-making processes in various fields such as robotics, AI, and complex system management. The ultimate objective is to contribute to the field of robotics and autonomous systems by providing new theoretical insights and practical tools that can adapt and evolve in response to changing conditions, thereby enabling more effective and strategic decision-making in uncertain scenarios.
Research project: 

My project is focused on constructing a dynamic, relationship-based knowledge graph. This graph intricately maps interactions and dependencies among various entities such as agents, objects, and subjects within a specific environment. A distinctive feature of this project is its ability to identify and integrate hidden subjects into the knowledge graph, enabling it to evolve as new data or entities are introduced. Central to this research is the application of game theory to create multiple simulations, around five in total, each projecting potential outcomes contingent on the actions of agents within the system. These simulations are crucial for envisioning various future scenarios and understanding the range of possible developments. Further enhancing the project's robustness is the use of model checking tools, like Rational Verification, to conduct probabilistic verification. This process is designed to identify the most viable simulations with high success probability. The overarching aim is to leverage these insights for improved decision-making and adaptive learning, enabling systems to refine their strategies in response to the dynamic changes in the environment. This project not only seeks to contribute significant advancements to the fields of game theory and adaptive learning but also aims to provide innovative methods for handling uncertainty and complexity in diverse settings.

About me: 

I am currently advancing my academic journey as a PhD student in Robotics and Autonomous Systems at the University of Edinburgh and Heriot-Watt University. With a strong foundation in computer programming and a decade of industry experience, my focus has shifted towards the cutting-edge field of AI and robotics. I hold an MSc in Artificial Intelligence from the University of St. Andrews, where I honed my skills in machine learning and deep learning. My research interests lie at the intersection of AI technologies and autonomous systems, aiming to develop innovative solutions that can navigate complex, real-world challenges. Beyond my academic pursuits, I bring a diverse set of skills and experiences, including my role as Project Manager & Software Engineer at Hatless Studios and Managing Director at Young Enterprise. Fluent in English and Tamil, I am deeply passionate about exploring the potential of AI in various domains, and I am committed to contributing meaningfully to the field through my research.

Student type: 
Current student