1 articles

TD-Gammon

Index of editorial tags and filtered coverage.

TD-Gammon: reinforcement learning that rivaled humans

Gerald Tesauro trained a TD-lambda network for backgammon, reaching elite level without massive hand-coded expertise.