Category: Cool Stuff
-
Unit-Vector Gradients: Experiments in Self-Organizing Geometry
These numerical experiments grew out of work on reinforcement-learning stability. They explore geometric patterns produced by gradient updates, with comparisons to physical structures used as exploratory analogies. Over the last several months I’ve been searching for the ultimate learning algorithm that could reliably learn strategic plays in adversarial, sparse reward settings like foosball, air hockey…
