Organization: UC Berkeley
Country: USA
Website: berkeley.edu
Year Founded: 1868
Number of Employees: 500+
Description:
UC Berkeley’s DayDreamer is a reinforcement-learning (RL) algorithm that, for example, taught a quadruped to walk in just one hour without interacting with a simulator. DayDreamer uses neural networks to interact with the environment, and then uses the information it gathers to learn a world model. This world model allows machine learning to predict the results of a series of actions. This predicted behavior is used with RL to train a controller for the robot.
UC Berkeley’s DayDreamer is a reinforcement-learning (RL) algorithm that, for example, taught a quadruped to walk in just one hour without interacting with a simulator. DayDreamer uses neural networks to interact with the environment, and then uses the information it gathers to learn a world model. This world model allows machine learning to predict the results of a series of actions. This predicted behavior is used with RL to train a controller for the robot.