{"pk":30044,"title":"Cross-Domain Adversarial Reprogramming of a Recurrent Neural Network","subtitle":null,"abstract":"Neural networks are vulnerable to adversarial attacks. These attacks can be untargeted, causing the model to make anyerror, or targeted, causing the model to make a specific error. Adversarial Reprogramming introduces a type of attackthat reprograms the network to perform an entirely new task from its original function. Additional inputs in a pre-trainednetwork can repurpose the network to a different task. Previous work has shown adversarial reprogramming possible insimilar domains, such as an image classification task in ImageNet being repurposed for CIFAR-10. A natural questionis whether such reprogramming is feasible across any task for neural networks a positive answer would have significantimpact both on wider applicability of ANNs, but also require rethinking their security. We attempt for the first timereprogramming across domains, repurposing a text classifier to an image classifier, using a recurrent neural network aprototypical example of a Turing universal network.","language":"eng","license":{"name":"","short_name":"","text":null,"url":""},"keywords":[],"section":"Poster Session 3","is_remote":true,"remote_url":"https://escholarship.org/uc/item/2dz4z9xv","frozenauthors":[{"first_name":"Alexandra","middle_name":"","last_name":"Proca","name_suffix":"","institution":"Massachusetts Institute of Technology","department":""},{"first_name":"Andrzej","middle_name":"","last_name":"Banburski","name_suffix":"","institution":"Massachusetts Institute of Technology","department":""},{"first_name":"Tomaso","middle_name":"","last_name":"Poggio","name_suffix":"","institution":"Massachusetts Institute of Technology","department":""}],"date_submitted":null,"date_accepted":null,"date_published":"2020-01-01T18:00:00Z","render_galley":null,"galleys":[{"label":"PDF","type":"pdf","path":"https://journalpub.escholarship.org/cognitivesciencesociety/article/30044/galley/19898/download/"}]}