Background Adversarial purification is a defense mechanism for safe-guarding classifiers against adversarial attacks without knowing the type of attacks or training of the classifier. These techniques analyze and eliminate adversarial perturbations from the attacked inputs, and help to restore purified samples that retain similarity to the attacked...

Background The rapid dissemination of news is an important factor for the global population to stay up to date with current events, or to make well-informed decisions in dynamic sectors like the stock market. Journalists typically include large amounts of quantitative information using data visualization to help enhance accessibility and comprehension...

Supply chains provide a network describing how resources move from a source to delivery for consumption by the end user. Supply chains also involve the components involved in production and transportation of the commodities being moved in the supply chain. There is a need for modeling links in a supply chain to estimate supply and demand and to ensure...

Federated learning (FL) is envisaged to be a promising solution for collaboratively training a machine learning model while keeping the training data decentralized and private. For example, in FL, different participants might have different samples of data from different classes. Some participants might have no samples from some of the subclasses...

Present day autonomous systems are cyber-physical systems (CPS) that involve multi-dimensional interaction between the cyber machine and the operational environment. One of the key aspects of these autonomous systems is that interaction patterns change over the lifetime of usage. Changes in CPS interaction patterns can be unintentional or intentional...

AI systems are increasingly being used in critical applications where safety and reliability are paramount such as health care, autonomous driving, and financial services. Numerous AI systems operate as “black boxes,” meaning a system that processes inputs and produces outputs without revealing how it reaches its conclusions. This makes...

Background In recent years, advancements in machine learning have revolutionized various domains, but concerns about bias and unfairness in automated decision-making systems have gained significant attention. One of the causes of unfairness in AI systems is the reliance on training data. Aleatoric uncertainty refers to the inherent randomness or variability...

Background Touchless interaction systems have grown in prevalence in recent years, including gesture-based interaction systems. Gesture recognition forms an integral part of human-computer interaction (HCI) technologies that deal with recognizing various movements of the human body, and interpreting them as commands to interact with a machine. Hand...

Many modern science and engineering applications, such as machine learning, hyperparameter optimization of neural networks, robotics, cyber-physical systems, etc., call for modeling techniques to model black-box functions. Gaussian Process (GP) modeling is a popular Bayesian non-parametric framework heavily employed to model expensive black-box functions...

Real-time monitoring and control of distribution networks was traditionally deemed unnecessary because it had radial configuration, unidirectional power flows, and predictable load patterns. However, the fast growth of behind-the-meter generation, particularly solar photovoltaic, electric vehicles, and storage, is transitioning the distribution system...