Detection Of Adversarial Attacks And Characterization Of Adversarial Subspace

This video program is a part of the Premium package:

Detection Of Adversarial Attacks And Characterization Of Adversarial Subspace


  • IEEE MemberUS $11.00
  • Society MemberUS $0.00
  • IEEE Student MemberUS $11.00
  • Non-IEEE MemberUS $15.00
Purchase

Detection Of Adversarial Attacks And Characterization Of Adversarial Subspace

0 views
  • Share
Create Account or Sign In to post comments
Adversarial attacks have always been a serious threat for any data-driven model. In this paper, we explore subspaces of adversarial examples in unitary vector domain, and we propose a novel detector for defending our models trained for environmental sound
Adversarial attacks have always been a serious threat for any data-driven model. In this paper, we explore subspaces of adversarial examples in unitary vector domain, and we propose a novel detector for defending our models trained for environmental sound