106
click to vote
ICML
2003
IEEE
16 years 4 months ago
2003
IEEE
A satisfactory multiagent learning algorithm should, at a minimum, learn to play optimally against stationary opponents and converge to a Nash equilibrium in self-play. The algori...