HackerMarI/O - Machine Learning for Video Games
SethBlingYouTube ↗
Further reading 2
Where does learning get a signal without an answer key? →Start with an unlabeled patch of an image, then switch to a game that requires a sequence of actions. Neither has an answer key, but finding structure and choosing a move need different signals.This video helps answer · How do game agents improve beyond their starting point?What improved when a game agent got better? →A rising score might reflect a better network, deeper search or a loophole. Trace what the agent observes, what it tries and why the resulting move earns a reward.This video helps answer · Can a network evolve its own wiring?
Caption notes
MarI/O evolves networks with NEAT, rewarding progress through a Mario level. Selection and mutation produce increasingly effective controller actions.
Original passages · 5
The white squares stand for blocks the player can stand onOnly the neural networks that produce the highest fitness are selected to be bredbuild up neural networks from barebones without presupposing the best structureOnly the neural networks that produce the highest fitness are selectedwithout presupposing the best structure for the neurons and their connections