Mark Moldwin | Searcharxiv

arXiv2019

Principles Of Heliophysics: a textbook on the universal processes behind planetary habitability

Heliophysics is the system science of the physical connections between the Sun and the solar system. As the physics of the local cosmos, it embraces space weather and planetary habitability. The wider view of comparative heliophysics forms a template for conditions in exoplanetary systems and provides a view over time of the aging Sun and its magnetic activity, of the heliosphere in different settings of the interstellar medium and subject to stellar impacts, of the space physics over evolving planetary dynamos, and of the long-term influence on planetary atmospheres by stellar radiation and wind. Based on a series of NASA-funded summer schools for early-career researchers, this textbook is intended for students in physical sciences in later years of their university training and for beginning graduate students in fields of solar, stellar, (exo-)planetary, and planetary-system sciences. The book emphasizes universal processes from a perspective that draws attention to what provides Earth (and similar (exo-)planets) with a relatively stable setting in which life as we know it could thrive. The text includes 200 "Activities" in the form of exercises, explorations, literature readings, "what if" challenges, and group discussion topics; many of the Activities provide additional information complementing the main text. Solutions and discussions are included in an Appendix for a selection of the exercises.

astro-ph.SR↗

arXiv2018

Simple Regret Minimization for Contextual Bandits

There are two variants of the classical multi-armed bandit (MAB) problem that have received considerable attention from machine learning researchers in recent years: contextual bandits and simple regret minimization. Contextual bandits are a sub-class of MABs where, at every time step, the learner has access to side information that is predictive of the best arm. Simple regret minimization assumes that the learner only incurs regret after a pure exploration phase. In this work, we study simple regret minimization for contextual bandits. Motivated by applications where the learner has separate training and autonomous modes, we assume that the learner experiences a pure exploration phase, where feedback is received after every action but no regret is incurred, followed by a pure exploitation phase in which regret is incurred but there is no feedback. We present the Contextual-Gap algorithm and establish performance guarantees on the simple regret, i.e., the regret during the pure exploitation phase. Our experiments examine a novel application to adaptive sensor selection for magnetic field estimation in interplanetary spacecraft, and demonstrate considerable improvement over algorithms designed to minimize the cumulative regret.

stat.ML↗