“If it works the first time its probably wrong”- I’ve been doomed to life of correction and striving for better since I first heard this.

The idea of systems that are able to learn, adapt and implement have fascinated me for as long as I can remember. This paired with an intrinsic love for sports and competition has led to a reality filled with yelling at and scrutinizing agents until they do better (results may vary)

I have found myself wrestling with the ideas of consciousness and optimality. To be specific, that it may require multiple agents to build a conscious. The agent you use when you feel hungry and seek food is different to the agent you use to produce a painting. All agents not necessarily disjoint, but in some sense independent.

This has manifested into projects with the goal of producing distinct behaviours within the same environment or for the same general task. My domain of choice being Pokemon. A perfect mixture of an extraordinarily large state space, needing distinct behaviours for the same tasks and appeal to my childhood.

A hence proclaim myself the Pokemon Master

Interests
  • Artificial Intelligence for Games
  • Reinforcement Learning
  • Representation Learning
Education
  • BSc (Hons) in Computer Science, 2024

    University of the Witwatersrand

  • BSc in Computer Science, 2023

    University of the Witwatersrand