On Regret-Optimal Learning in Decentralized Multiplayer Multiarmed Bandits | IEEE Journals & Magazine | IEEE Xplore