UNICAD: A unified approach for attack detection, noise reduction and novel class identification
As the use of Deep Neural Networks (DNNs) becomes pervasive, their vulnerability to
adversarial attacks and limitations in handling unseen classes poses significant challenges.
The state-of-the-art offers discrete solutions aimed to tackle individual issues covering
specific adversarial attack scenarios, classification or evolving learning. However, real-world
systems need to be able to detect and recover from a wide range of adversarial attacks
without sacrificing classification accuracy and to flexibly act in unseen scenarios. In this …
adversarial attacks and limitations in handling unseen classes poses significant challenges.
The state-of-the-art offers discrete solutions aimed to tackle individual issues covering
specific adversarial attack scenarios, classification or evolving learning. However, real-world
systems need to be able to detect and recover from a wide range of adversarial attacks
without sacrificing classification accuracy and to flexibly act in unseen scenarios. In this …
UNICAD: A Unified Approach for Attack Detection, Noise Reduction and Novel Class Identification
A Lopez Pellicer, K Giatgong, Y Li, N Suri… - arXiv e …, 2024 - ui.adsabs.harvard.edu
As the use of Deep Neural Networks (DNNs) becomes pervasive, their vulnerability to
adversarial attacks and limitations in handling unseen classes poses significant challenges.
The state-of-the-art offers discrete solutions aimed to tackle individual issues covering
specific adversarial attack scenarios, classification or evolving learning. However, real-world
systems need to be able to detect and recover from a wide range of adversarial attacks
without sacrificing classification accuracy and to flexibly act in {\bf unseen} scenarios. In this …
adversarial attacks and limitations in handling unseen classes poses significant challenges.
The state-of-the-art offers discrete solutions aimed to tackle individual issues covering
specific adversarial attack scenarios, classification or evolving learning. However, real-world
systems need to be able to detect and recover from a wide range of adversarial attacks
without sacrificing classification accuracy and to flexibly act in {\bf unseen} scenarios. In this …
Showing the best results for this search. See all results
