In a new paper published last week, scientists at AI research lab DeepMind claimed to have taken the “first steps to train an agent capable of playing many different games without needing human interaction data.” Their new AI system, which they describe as “open-ended learning,” includes a 3D environment with realistic dynamics and deep reinforcement learning agents that can learn to solve a wide range of challenges.
DeepMind's new reinforcement learning system
DeepMind's new reinforcement learning system
DeepMind's new reinforcement learning system
In a new paper published last week, scientists at AI research lab DeepMind claimed to have taken the “first steps to train an agent capable of playing many different games without needing human interaction data.” Their new AI system, which they describe as “open-ended learning,” includes a 3D environment with realistic dynamics and deep reinforcement learning agents that can learn to solve a wide range of challenges.