CVSL Logo
FrancaisHome
AboutPeopleResearchPublicationsEventsProfile
About
Seminars


The CerVIM Seminars, Université Laval are held on Fridays at 11:00 a.m.
Please see the program for more details.

 

 

 

CERVIM

REPARTI

MIVIM

13-04-2018

Mahdieh Abbasi
Laboratoire LVSN
Dép. de génie électrique et de génie informatique, Université Laval

Towards Dependable Deep Convolutional Neural Networks (CNNs) with Out-distribution Learning



Abstract

The easiness at which adversarial instances can be generated in deep neural networks raises some fundamental questions on their functioning and concerns on their use in critical systems. In this presentation, we draw a connection between over-generalization and adversaries: a possible cause of adversaries lies in models designed to make decisions all over the input space, leading to inappropriate high-confidence decisions in parts of the input space not represented in the training set. We empirically show that an augmented CNN, which is not trained on any types of adversaries, can increase the robustness by either rejecting or classifying correctly most adversarial examples generated using some well-known attack methods, without sacrificing the accuracy of the augmented CNN on clean samples significantly.


The seminar will be presented at 11:30 a.m. in room PLT-1120.




     
   
   

©2002-. Computer Vision and Systems Laboratory. All rights reserved