Malware remains a big threat to cyber security, calling for machine learning based malware
detection. While promising, such detectors are known to be vulnerable to evasion attacks …

We show that classifiers trained with random region proposals achieve state-of-the-art Open-
world Object Detection (OWOD): they can not only maintain the accuracy of the known …

While deep neural networks have been achieving state-of-the-art performance across a
wide variety of applications, their vulnerability to adversarial attacks limits their widespread …

Similarity metrics have played a significant role in computer vision to capture the underlying
semantics of images. In recent years, advanced similarity metrics, such as the Learned …

Abstract Vision Transformers (ViTs) have emerged as a compelling alternative to
Convolutional Neural Networks (CNNs) in the realm of computer vision showcasing …

We consider sensitivity of a generic stochastic optimization problem to model uncertainty.
We take a non-parametric approach and capture model uncertainty using Wasserstein balls …

The worst-case training principle that minimizes the maximal adversarial loss, also known as
adversarial training (AT), has shown to be a state-of-the-art approach for enhancing …

ABSTRACT Deep Neural Networks (DNNs) have demonstrated to be powerful machine
learning models for approximating complex functions. In this work, we provide an exact …

Adversarial training (AT) refers to integrating adversarial examples--inputs altered with
imperceptible perturbations that can significantly impact model predictions--into the training …

The robustness of image segmentation has been an important research topic in the past few
years as segmentation models have reached production-level accuracy. However, like …