Learning Soccer Juggling Skills With Layer-wise Mixture of Experts
Event Type
Technical Paper
Interest Areas
Research & Education
Presentation Types
Registration Categories
Full Conference Supporter
Full Conference
Virtual Conference Supporter
Virtual Conference
Exhibitor Additional Full Conference
Exhibitor Full Conference
DescriptionWe introduce a method for learning physics-based soccer juggling skills via deep reinforcement learning. Innovations include a Layer-wise Mixture of Experts neural network policy for efficient learning, a control graph for authoring the many skills and their transitions, and an adaptive random walk curriculum on the control graph.