Presentation
Learning Soccer Juggling Skills With Layer-wise Mixture of Experts
Event Type
Technical Paper
Animation/Simulation
Research & Education
In Person
Full Conference Supporter
Full Conference
Exhibitor Additional Full Conference
Exhibitor Full Conference
This session WILL NOT be recorded.
TimeTuesday, 9 August 20222:38pm - 2:43pm PDT
LocationEast Building, Ballroom A/B
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.
