English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Temporal variations of apparent magnitude, called light curves, areobservational statistics of interest captured by telescopes over long periodsof time. Light curves afford the exploration of Space Domain Awareness (SDA)objectives such as object identification or pose estimation as latent variableinference problems. Ground-based observations from commercial off the shelf(COTS) cameras remain inexpensive compared to higher precision instruments,however, limited sensor availability combined with noisier observations canproduce gappy time-series data that can be difficult to model. These externalfactors confound the automated exploitation of light curves, which makes lightcurve prediction and extrapolation a crucial problem for applications.Traditionally, image or time-series completion problems have been approachedwith diffusion-based or exemplar-based methods. More recently, Deep NeuralNetworks (DNNs) have become the tool of choice due to their empirical successat learning complex nonlinear embeddings. However, DNNs often require largetraining data that are not necessarily available when looking at uniquefeatures of a light curve of a single satellite.  In this paper, we present a novel approach to predicting missing and futuredata points of light curves using Gaussian Processes (GPs). GPs are non-linearprobabilistic models that infer posterior distributions over functions andnaturally quantify uncertainty. However, the cubic scaling of GP inference andtraining is a major barrier to their adoption in applications. In particular, asingle light curve can feature hundreds of thousands of observations, which iswell beyond the practical realization limits of a conventional GP on a singlemachine. Consequently, we employ MuyGPs, a scalable framework forhyperparameter estimation of GP models that uses nearest neighborssparsification and local cross-validation. MuyGPs...

Light curve completion and forecasting using fast and scalable Gaussian processes (MuyGPs)