New method uses crowdsourced feedback to help train robots

A new technique enables an AI agent to be guided by data crowdsourced asynchronously from nonexpert human users as it learns to complete a task through reinforcement learning. The method trains the robot faster and better than other approaches.