DeepCloak: Masking Deep Neural Network Models for Robustness Against Adversarial Samples
This addresses security risks in DNN systems, particularly in sensitive settings, but appears incremental as it builds on existing defensive approaches.
The paper tackles the vulnerability of deep neural networks to adversarial samples by introducing DeepCloak, a defensive mechanism that removes unnecessary features to limit attack capacity, resulting in increased robustness as shown in experimental results.
Recent studies have shown that deep neural networks (DNN) are vulnerable to adversarial samples: maliciously-perturbed samples crafted to yield incorrect model outputs. Such attacks can severely undermine DNN systems, particularly in security-sensitive settings. It was observed that an adversary could easily generate adversarial samples by making a small perturbation on irrelevant feature dimensions that are unnecessary for the current classification task. To overcome this problem, we introduce a defensive mechanism called DeepCloak. By identifying and removing unnecessary features in a DNN model, DeepCloak limits the capacity an attacker can use generating adversarial samples and therefore increase the robustness against such inputs. Comparing with other defensive approaches, DeepCloak is easy to implement and computationally efficient. Experimental results show that DeepCloak can increase the performance of state-of-the-art DNN models against adversarial samples.