We examine the general problem of inter-domain Gaussian Processes (GPs): problems where the GP realization and the noisy observations of that realization lie on different domains. When the mapping between those domains is linear, such as integration or differentiation, inference is still closed form. However, many of the scaling and approximation techniques that our community has developed do not apply to this setting. In this work, we introduce the hierarchical inducing point GP (HIP-GP), a scalable inter-domain GP inference method that enables us to improve the approximation accuracy by increasing the number of inducing points to the millions. HIP-GP, which relies on inducing points with grid structure and a stationary kernel assumption, is suitable for low-dimensional problems. In developing HIP-GP, we introduce (1) a fast whitening strategy, and (2) a novel preconditioner for conjugate gradients which can be helpful in general GP settings.

Once shared, our personal information on the Internet is no longer private. We routinely receive emails from companies that we have not had any known interaction with, and are receiving an increasingly large volume of spam phone calls. In this paper, we describe interim results from an experiment designed to quantify who is using and distributing our personally identifying information (PII). To do this, we set up 300 fake identities, each with an email address and around half with a live phone number, and performed one-time online interactions with 188 distinct companies. Over a 9-month span, we received around 20,000 artifacts and found that reputable companies, surprisingly, do not sell our information in ways that we could detect, that there was no observation of undue foreign interest during the election, and that the classic “extended vehicle warranty” scam is still in active use today.

Online interactions typically require a user to input personally identifiable information (PII) such as their name, email, and demographic characteristics. The service provider may then use that PII to send correspondence to the user’s email address or phone number, either through themselves or a third party. This study aims to create a tentative framework for measuring political bias within PII-harnessing communications. Three distinct spheres of analysis (time, corporate political values, and foreign senders) are utilized to develop this system of measurement. Although the results of a small-scale test of our method were inconclusive, our process for quantitatively measuring political bias nonetheless serves as a proof of concept that can be applied to future research.

Abstract We present a state‐of‐the‐art report on visualization in astrophysics. We survey representative papers from both astrophysics and visualization and provide a taxonomy of existing approaches based on data analysis tasks. The approaches are classified based on five categories: data wrangling, data exploration, feature identification, object reconstruction, as well as education and outreach. Our unique contribution is to combine the diverse viewpoints from both astronomers and visualization experts to identify challenges and opportunities for visualization in astrophysics. The main goal is to provide a reference point to bring modern data analysis and visualization techniques to the rich datasets in astrophysics.