Robots play Hide and Seek to learn dynamic interaction.
Robots play Hide and Seek to learn dynamic interaction. Seeker and Hider use vision models + RL policies to change hiding strategies based on live feedback.
