Search for: All records

Since SSH’s standardization nearly 20 years ago, real-world requirements for a remote access protocol and our understanding of how to build secure cryptographic network protocols have both evolved significantly. In this work, we introduce Hop, a transport and remote access protocol designed to support today’s needs. Building on modern cryptographic advances, Hop reduces SSH protocol complexity and overhead while simultaneously addressing many of SSH’s shortcomings through a cryptographically-mediated delegation scheme, native host identification based on lessons from TLS and ACME, client authentication for modern enterprise environments, and support for client roaming and intermittent connectivity. We present concrete design requirements for a modern remote access protocol, describe our proposed protocol, and evaluate its performance. We hope that our work encourages discussion of what a modern remote access protocol should look like in the future. 

Abstract We present a general class of machine learning algorithms called parametric matrix models. In contrast with most existing machine learning models that imitate the biology of neurons, parametric matrix models use matrix equations that emulate physical systems. Similar to how physics problems are usually solved, parametric matrix models learn the governing equations that lead to the desired outputs. Parametric matrix models can be efficiently trained from empirical data, and the equations may use algebraic, differential, or integral relations. While originally designed for scientific computing, we prove that parametric matrix models are universal function approximators that can be applied to general machine learning problems. After introducing the underlying theory, we apply parametric matrix models to a series of different challenges that show their performance for a wide range of problems. For all the challenges tested here, parametric matrix models produce accurate results within an efficient and interpretable computational framework that allows for input feature extrapolation. 

Differential privacy is the dominant standard for formal and quantifiable privacy and has been used in major deployments that impact millions of people. Many differentially private algorithms for query release and synthetic data contain steps that reconstruct answers to queries from answers to other queries that have been measured privately. Reconstruction is an important subproblem for such mecha- nisms to economize the privacy budget, minimize error on reconstructed answers, and allow for scalability to high-dimensional datasets. In this paper, we introduce a principled and efficient postprocessing method ReM (Residuals-to-Marginals) for reconstructing answers to marginal queries. Our method builds on recent work on efficient mechanisms for marginal query release, based on making measurements using a residual query basis that admits efficient pseudoinversion, which is an important primitive used in reconstruction. An extension GReM-LNN (Gaussian Residuals-to-Marginals with Local Non-negativity) reconstructs marginals under Gaussian noise satisfying consistency and non-negativity, which often reduces error on reconstructed answers. We demonstrate the utility of ReM and GReM-LNN by applying them to improve existing private query answering mechanisms. 

Abstract Severe convective storms and tornadoes rank among nature’s most hazardous phenomena, inflicting significant property damage and casualties. Near-surface weather conditions are closely governed by large-scale synoptic patterns. It is crucial to delve into the involved multiscale associations to understand tornado potential in response to climate change. Using clustering analysis, this study unveils that leading synoptic patterns driving tornadic storms and associated spatial trends are distinguishable across geographic regions in the U.S. Synoptic patterns with intense forcing featured by intense upper-level eddy kinetic energy and a dense distribution of Z500 fields dominate the increasing trend in tornado frequency in the southeast U.S., generating more tornadoes per event. Conversely, the decreasing trend noted in certain regions of the central Great Plains is associated with weak upper-level synoptic forcing. These findings offer an explanation of observational changes in tornado occurrences, suggesting that the physical mechanisms driving those changes differ across regions.