Weighted Voting Via No-Regret Learning