相关论文: Pluto's Surface Mapping using Unsupervised Learnin…
We have developed a method that maps large astronomical images onto a two-dimensional map and clusters them. A combination of various state-of-the-art machine learning (ML) algorithms is used to develop a fully unsupervised image quality…
Pluto's landscape is shaped by the cycles of the volatile ices covering its surface. In particular, the Sputnik Planitia (SP) ice sheet displays a large diversity of terrains, with bright and dark plains, pits, topographic depressions and…
Context: Pluto possesses a thin atmosphere, primarily composed of nitrogen, in which the detection of methane has been reported. Aims: The goal is to constrain essential but so far unknown parameters of Pluto's atmosphere such as the…
We introduce a novel learning-based method for encoding and manipulating 3D surface meshes. Our method is specifically designed to create an interpretable embedding space for deformable shape collections. Unlike previous 3D mesh…
We propose a novel unsupervised framework for \emph{Invariant Risk Minimization} (IRM), extending the concept of invariance to settings where labels are unavailable. Traditional IRM methods rely on labeled data to learn representations that…
The principal goal of the NASA Terrestrial Planet Finder (TPF) and ESA Darwin mission concepts is to directly detect and characterize extrasolar terrestrial (Earth-sized) planets. This first generation of instruments is expected to provide…
Digital co-addition of astronomical images is a common technique for increasing signal-to-noise and image depth. A modification of this simple technique has been applied to the detection of minor bodies in the Solar System: first stationary…
I present a strategy for unsupervised manifold learning on local atomic environments in molecular simulations based on simple rotation- and permutation-invariant three-body features. These features are highly descriptive, generalize to…
This paper illustrates an application of machine learning (ML) within a complex system that performs grade estimation. In surface mining, assay measurements taken from production drilling often provide useful information that allows…
Airborne hyperspectral images can be used to map the land cover in large urban areas, thanks to their very high spatial and spectral resolutions on a wide spectral domain. While the spectral dimension of hyperspectral images is highly…
Land use/land cover change (LULC) maps are integral resources in earth science and agricultural research. Due to the nature of such maps, the creation of LULC maps is often constrained by the time and human resources necessary to accurately…
Map-to-map matching is a critical task for aligning spatial data across heterogeneous sources, yet it remains challenging due to the lack of ground truth correspondences, sparse node features, and scalability demands. In this paper, we…
A link between exposures of water (H${}_{2}$O) ice with traces of an ammoniated compound (e.g., a salt) and the probable effusion of a water-rich cryolava onto the surface of Pluto has been established in previous investigations (Dalle Ore…
The first exploration of Pluto was motivated by (i) the many intriguing aspects of this body, its atmosphere, and its giant impact binary-planet formation; as well as (ii) the scientific desire to initiate the reconnaissance of the…
We develop a new retrieval scheme for obtaining two-dimensional surface maps of exoplanets from scattered light curves. In our scheme, the combination of the L1-norm and Total Squared Variation, which is one of the techniques used in sparse…
Context. Filaments are ubiquitous in the Galaxy, and they host star formation. Detecting them in a reliable way is therefore key towards our understanding of the star formation process. Aims. We explore whether supervised machine learning…
We study the release of various elements from Callisto's surface into its exosphere by plasma sputtering. The cold Jovian plasma is simulated with a 3D plasma-planetary interaction hybrid model, which produces 2D surface precipitation maps…
We develop a new analysis method that allows us to compare multi-dimensional observables to a theoretical model. The method is based on unsupervised clustering algorithms which assign the observational and simulated data to clusters in high…
We review the importance of recent UV observations of solar system targets and discuss the need for further measurements, instrumentation and laboratory work in the coming decade. In the past decade, numerous important advances have been…
The recent discovery of two large trans-Neptunian objects (TNOs) 2003 UB313 and 2005 FY9, with surface properties similar to those of Pluto, provides an exciting new laboratory for the study of processes considered for Pluto and Triton:…