Adversarially Robust Policy Learning: Active construction of physically-plausible perturbations.

Ajay Mandlekar, Yuke Zhu, Animesh Garg, Li Fei-Fei, Silvio Savarese

Browse the full IROS paper archive.