Image:Google DeepMind(X) Google DeepMind は深層強化学習(Deep RL)を使用して、小さな二足歩行ロボットにサッカーを教え、そのプレイする様子をX(Twitter)に投稿した。 Robotics has long used games like soccer to test complex motion, coordination and environment awareness. 🤖 Our agents were trained in simulation using the MuJoCo physics engine, and then transferred to small humanoid robots with 20 actuated joints. https://t.co/RFBxLG7qBV — Google DeepMin