I am currently a MSc student at the University of the Witwatersrand, South Africa. I am also a practising Quantitative Analyst working in the field of Model Validation. I have been drawn to the field of Artificial Intelligence and Robotics mostly out of curiosity, but also from recent developments in the financial services industry making use of this technology to improve efficiency and performance. Outside of work and studies I am an avid gamer, hiker, and Rubik’s Cube speedsolver.
The research question I am aiming to answer is whether it is possible for an agent to purposely create information asymmetry in a game in a way that it is able to exploit later. This would require an agent to be able to estimate its own skill, it’s opponents skill, and its opponent’s estimation of the agent’s skill, and then choose actions in a way that lower the opponent’s estimation of the agent’s skill below the true estimate of the agent’s skill.