Claire Glanois

Claire Glanois

Postdoc

IT University of Copenhagen

I am a mathematician, previously working at the intersection of number theory and algebraic geometry. My current research is looking at how agents may learn, adapt, self-organise, and learn to learn, in a more open-ended way. I am also interested in making AI algorithms more interpretable, notably through neuro-logic approaches. In parallel, I am sensitive to algorithmic.{plurality, normativity, transparency}, and experiment with ways for artificial agents to be more diverse, situated, and homegrown.

Interests
  • Deep Reinforcement Learning
  • Open-Endedness
  • Self-Organisation
  • Neuro-Logic
  • Interpretable AI
  • Creative AI
  • Continual Learning
  • Number Theory
  • Algebraic Geometry