Enhancing video game AI via machine learning and evolutionary strategic interactions

Project summary

• Investigate current research in game artificial intelligence (AI) and games technology for the creation of more intelligent, engaging human-like AI agents. The purpose is to understand the link between knowledge representation and strategic decision-making
• Design a framework to implement human-like AI agents and measure their accuracy for cooperation and collective decision-making. A well-known game engine such as Unity or Unreal Engine will be used to develop this framework. The developed framework will provide an experimental setup to extend current research to artificial general intelligence (AGI)
• Evaluate the performance of the proposed methods with current state-of-the-art implementations. Extend the research work to address societal and scientific challenges beyond games, eg in education

The first developments in artificial intelligence have used several games such as chess or Go as the ideal test bed, with the aim to produce algorithms to outsmart their human counterpart. In a similar way, video games have recently gained popularity for their applications in the AI research community, and as an efficient way to collect data. However, most of the academic work in the area of AI focused on traditional board and card games where limited AI techniques have been tested. In order to overcome this gap in the academic research literature, a framework for research on game AI will be developed within this project.

The game framework will integrate elements of machine learning and game theory to create engaging AI agents that are self-aware and able to interact strategically with the surrounding environment, the human players and among themselves. Many approaches and algorithms used recently in this area include Self-Adaptive Monte Carlo Tree Search or Deep Reinforcement Learning when only the rules are available as input. The emphasis in this project will be placed in understanding an efficient way to represent knowledge in order to develop human-like AI for cooperation and competition in a variety of contexts, eg collective decision-making.

Research centre

Data Science Research Centre 

Entry requirements

Applicants must hold a minimum of an Upper-Second (2:1) UK Honours degree or international equivalent in computer science, engineering or related subject. Applicants normally hold a masters degree, although we consider applications from exceptional candidates with a fitting undergraduate degree.

It is an essential requirement to have excellent programming skills in C++, C# and/or other relevant programming languages used in game development as well as a proven experience with one or more game engines, including Unity and Unreal Engine.

International students may also need to meet our English language requirements. Find out more about our entry requirements for international students.

Project specific requirements must align with the University’s standard requirements.

How to apply

Please contact Dr Leonardo Stella (l.stella@derby.ac.uk) in the first instance for more information on how to apply.

The University has four starting points each year for MPhil/PhD programmes (September, January, March and June). Applications should be made at least three months before you would want to start your programme. Please note that, if you require a visa, additional time will be required.

Funding

Self-funded by student. There is a range of options that may be available to you to help you fund your PhD.

Supervisors

References

L. Stella and D. Bauso, “Bio-inspired Evolutionary Dynamics on Complex Networks under Uncertain Cross-inhibitory Signals”, Automatica, vol. 100, pp. 61-66, 2019

L. Stella and D. Bauso, “Bio-Inspired Evolutionary Game Dynamics in Symmetric and Asymmetric Models”, IEEE Control Systems Letters, vol. 2, no. 3, pp. 405-410, 2018

Y. Sun, B. Yuan, T. Zhang, B. Tang, W. Zheng and X. Zhou, “Research and Implementation of Intelligent Decision Based on a Priori Knowledge and DQN Algorithms in Wargame Environment”, Electronics, vol. 9, no. 10, pp. 1668, 2020

Y. Sun, B. Yuan, Z. Yongliang, Z. Wangwen, X. Qingfeng, T. Bojian and Z. Xianzhong, “Research on Action Strategies and Simulations of DRL and MCTS-based Intelligent Round Game”, International Journal of Control, Automation and Systems, 2021 (in press)

G. Yannakakis and J. Togelius, Artificial Intelligence and Games. Springer, 2018