no code implementations • 22 Feb 2020 • Kirthi Shankar Sivamani, Rajeev Sahay, Aly El Gamal
In this letter, we propose a novel method to detect adversarial inputs, by augmenting the main classification network with multiple binary detectors (observer networks) which take inputs from the hidden layers of the original network (convolutional kernel outputs) and classify the input as clean or adversarial.
no code implementations • 10 Dec 2019 • Kirthi Shankar Sivamani
We exploit the noisy feature maps by using an additional subnetwork to extract image feature maps and train an auto-encoder on perceptual losses of these feature maps.
no code implementations • 8 Jul 2019 • Kirthi Shankar Sivamani
A structured similarity index (SSIM) loss is used to enforce label retention while augmenting the training set.