Goto

Collaborating Authors

 Dixon, Peter


List and Certificate Complexities in Replicable Learning

arXiv.org Artificial Intelligence

Replicability and reproducibility in science are critical concerns. The fundamental requirement that scientific results and experiments be replicable/reproducible is central to the development and evolution of science. In recent years, these concerns have grown as several scientific disciplines turn to data-driven research, which enables exponential progress through data democratization and affordable computing resources. The replicability issue has received attention from a wide spectrum of entities, from general media publications (for example, The Economist's "How Science Goes Wrong," 2013 [eco13]) to scientific publication venues (for example, see [JP05, Bak16]) to professional and scientific bodies such as the National Academy of Sciences, Engineering, and Medicine (NASEM). The emerging challenges to replicability and reproducibility have been discussed in depth by a consensus study report published by NASEM [NAS19]. A broad approach taken to ensure the reproducibility/replicability of algorithms is to make the datasets, algorithms, and code publicly available.


Interstellar Object Accessibility and Mission Design

arXiv.org Artificial Intelligence

Abstract--Interstellar objects (ISOs) are fascinating and underexplored be best implemented using small spacecraft. The unification of celestial objects, providing physical laboratories to ISO detection, orbit characterization, and cruise trajectory with understand the formation of our solar system and probe the learning-based G&C algorithms for accurate low-V flybys composition and properties of material formed in exoplanetary represents a nearly end-to-end simulation and assessment of a systems. The recent Planetary Science and Astrobiology mission to visit an ISO. This process is simulated using JPL's Decadal Survey emphasized that a dedicated mission to an interstellar SmallSat Development Testbed, which determines the feasibility object would have high scientific value. A dedicated ISOs with varying characteristics, including a discussion of state spacecraft could resolve the shape, rotation properties, surface covariance estimation over the course of a cruise, handoffs from morphology, and composition of an asteroid-like ISO. Mass traditional navigation approaches to novel autonomous navigation spectroscopy techniques can probe the gas composition of a for fast flyby regimes, and overall recommendations about comet-like ISO.