401 citations · 1.3k across the 12 of their papers we have counts for
3 papers · 1 filter
Adversarial Patch
Tom B. Brown, Dandelion Mané, Aurko Roy +2
We present a method to create universal, robust, targeted adversarial image patches in the real world. The patches are universal because they can be used to attack any scene, robus…
Reinforcement Learning under Model Mismatch
Aurko Roy, Huan Xu, Sebastian Pokutta
We study reinforcement learning under model misspecification, where we do not have access to the true environment but only to a reasonably close approximation to it. We address thi…
Learning to Remember Rare Events
Łukasz Kaiser, Ofir Nachum, Aurko Roy +1
Despite recent advances, memory-augmented deep neural networks are still limited when it comes to life-long and one-shot learning, especially in remembering rare events. We present…