Related papers: Using Bayesian Deep Learning to infer Planet Mass …

One of the goals of current particle physics research is to obtain evidence for new physics, that is, physics beyond the Standard Model (BSM), at accelerators such as the Large Hadron Collider (LHC) at CERN. The searches for new physics are…

We introduce a new machine learning based technique to detect exoplanets using the transit method. Machine learning and deep learning techniques have proven to be broadly applicable in various scientific research areas. We aim to exploit…

We hypothesise that dust rings in protoplanetary discs formed by an embedded planet should have properties that reflect the planet's mass. We use 2D hydrodynamical simulations of planet-disc interactions to investigate this, focusing on…

Humans naturally perceive a 3D scene in front of them through accumulation of information obtained from multiple interconnected projections of the scene and by interpreting their correspondence. This phenomenon has inspired artificial…

Understanding how dust evolves in protoplanetary disks is crucial to constraining the initial conditions of planet formation. The apparent "mass budget problem", which stems from the comparison of the observed disk masses to the ones…

A comprehensive artificial intelligence system needs to not only perceive the environment with different `senses' (e.g., seeing and hearing) but also infer the world's conditional (or even causal) relations and corresponding uncertainty.…

Depth estimation is of critical interest for scene understanding and accurate 3D reconstruction. Most recent approaches in depth estimation with deep learning exploit geometrical structures of standard sharp images to predict corresponding…

Planet formation models rely on knowledge of the physical conditions and evolutionary processes in protoplanetary disks, in particular the grain size distribution and dust growth timescales. In theoretical models, several barriers exist…

Bayesian neural networks provide a direct and natural way to extend standard deep neural networks to support probabilistic deep learning through the use of probabilistic layers that, traditionally, encode weight (and bias) uncertainty. In…

Hydrodynamical simulations play a fundamental role in modern cosmological research, serving as a crucial bridge between theoretical predictions and observational data. However, due to their computational intensity, these simulations are…

Deep neural networks enable learning directly on the data without the domain knowledge needed to construct a feature set. This approach has been extremely successful in almost all machine learning applications. We propose a new framework…

This paper addresses the scalability problem of Bayesian deep neural networks. The performance of deep neural networks is undermined by the fact that these algorithms have poorly calibrated measures of uncertainty. This restricts their…

Fast and accurate magnitude prediction is the key to the success of earthquake early warning. We have proposed a new approach based on deep learning for P-wave magnitude prediction (EEWNet), which takes time series data as input instead of…

Large-scale vortices in protoplanetary disks are thought to form and survive for long periods of time. Hence, they can significantly change the global disk evolution and particularly the distribution of the solid particles embedded in the…

We predict magnitudes for young planets embedded in transition discs, still affected by extinction due to material in the disc. We focus on Jupiter-size planets at a late stage of their formation, when the planet has carved a deep gap in…

Time domain astronomy has emerged as a vibrant research field in recent years, focusing on celestial objects that exhibit variable magnitudes or positions. Given the urgency of conducting follow-up observations for such objects, the…

The similar orbital distances and detection rates of debris disks and the prominent rings observed in protoplanetary disks suggest a potential connection between these structures. We explore this connection with new calculations that follow…

In context of the universal presence of defects in additively manufactured (AM) metals, efficient computational tools are required to rapidly screen AM microstructures for mechanical integrity. To this end, a deep learning approach is used…

Being capable of estimating the pose of uncooperative objects in space has been proposed as a key asset for enabling safe close-proximity operations such as space rendezvous, in-orbit servicing and active debris removal. Usual approaches…

This paper presents an innovative deep learning pipeline which estimates the relative pose of a spacecraft by incorporating the temporal information from a rendezvous sequence. It leverages the performance of long short-term memory (LSTM)…