The Arcade Learning Environment: An Evaluation Platform for General Agents (2012, background): “Pong has been studied in a variety of contexts as an interesting reinforcement learning domain (Cobo et al., 2011; Stober & Kuipers, 2008; Monroy, Stanley, & Miikkulainen, 2006). The Atari 2600 Pong, however, is significantly more complex than Pong domains developed for research. Games can easily la”
Reinforcement learning from simultaneous human and MDP reward (2012, background): “Other recent work has employed non-expert humans to aid RL algorithms through guidance on feature selection [2], identification and demonstration of high-level actions [14], and giving natural language advice [9].”
A Neuroevolution Approach to General Atari Game Playing (2014, unknown): “Research in this area has focused on learning to play single games using perceptual imagery [52], human demonstration [9], discovery of objective functions [48], and use of raw visual inputs [30], [34], [35], [49].”