Retrieving exoplanet atmospheric parameters using random forest regression

Understanding of exoplanet atmospheres can be extracted from the transmission spectra using an important tool based on a retrieval technique. However, the traditional retrieval method (e.g. MCMC and nested sampling) consumes a lot of computational time. Therefore, this work aims to apply the random forest regression, one of the supervised machine learning technique, to retrieve exoplanet atmospheric parameters from the transmission spectra observed in the optical wavelength. We discovered that the random forest regressor had the best accuracy in predicting planetary radius ( R F i t 2 = 0.999) as well as acceptable accuracy in predicting planetary mass, temperature, and metallicity of planetary atmosphere. Our results suggested that the random forest regression consumes significantly less computing time while gives the predicted results equivalent to those of the nested sampling PLATON retrieval.

Nested sampling for materials

We review the materials science applications of the nested sampling (NS) method, which was originally conceived for calculating the evidence in Bayesian inference. We describe how NS can be adapted to sample the potential energy surface (PES) of atomistic systems, providing a straightforward approximation for the partition function and allowing the evaluation of thermodynamic variables at arbitrary temperatures. After an overview of the basic method, we describe a number of extensions, including using variable cells for constant pressure sampling, the semi-grand-canonical approach for multicomponent systems, parallelizing the algorithm, and visualizing the results. We cover the range of materials applications of NS from the past decade, from exploring the PES of Lennard–Jones clusters to that of multicomponent condensed phase systems. We highlight examples how the information gained via NS promotes the understanding of materials properties through a novel way of visualizing the PES, identifying thermodynamically relevant basins, and calculating the entire pressure–temperature(–composition) phase diagram. Graphic abstract

Optimális neurális hálózat kiválasztása Bayes-becslés segítségével

Ezen munkánkban célunk, hogy neurális hálózatokra alkalmazva a Bayes-becslést az \textit{a posteriori} becslések során a különböző modellek közül kiválasszuk a tanító adatoknak legjobban megfelelőt. Mindehhez egy sokdimenziós integrál kiszámítása szükséges, amely a hagyományos Monte-Carlo módszerekkel is nehéz feladat; erre a célra a {beágyazott mintavételezés (nested sampling)} algoritmust alkalmazzuk, és a számítások járulékos eredményeként kapjuk meg a betanított hálózatot a hiperparaméterek terében is bolyongást végezve. Továbbá rámutatunk arra, hogyan lehet ötvözni a gradiens visszaterjesztéses és a véletlen bolyongásos tanítást hibrid hálózatokat nyerve.

Nested sampling with plateaus

ABSTRACT It was recently emphasized that in the presence of plateaus in the likelihood function nested sampling (NS) produces faulty estimates of the evidence and posterior densities. After informally explaining the cause of the problem, we present a modified version of NS that handles plateaus and can be applied retrospectively to NS runs from popular NS software using anesthetic. In the modified NS, live points in a plateau are evicted one by one without replacement, with ordinary NS compression of the prior volume after each eviction but taking into account the dynamic number of live points. The live points are replenished once all points in the plateau are removed. We demonstrate it on a number of examples. Since the modification is simple, we propose that it becomes the canonical version of Skilling’s NS algorithm.

Microseismic event detection in large heterogeneous velocity models using Bayesian multimodal nested sampling

In passive seismic and microseismic monitoring, identifying and characterizing events in a strong noisy background is a challenging task. Most of the established methods for geophysical inversion are likely to yield many false event detections. The most advanced of these schemes require thousands of computationally demanding forward elastic-wave propagation simulations. Here we train and use an ensemble of Gaussian process surrogate meta-models, or proxy emulators, to accelerate the generation of accurate template seismograms from random microseismic event locations. In the presence of multiple microseismic events occurring at different spatial locations with arbitrary amplitude and origin time, and in the presence of noise, an inference algorithm needs to navigate an objective function or likelihood landscape of highly complex shape, perhaps with multiple modes and narrow curving degeneracies. This is a challenging computational task even for state-of-the-art Bayesian sampling algorithms. In this paper, we propose a novel method for detecting multiple microseismic events in a strong noise background using Bayesian inference, in particular, the Multimodal Nested Sampling (MultiNest) algorithm. The method not only provides the posterior samples for the 5D spatio-temporal-amplitude inference for the real microseismic events, by inverting the seismic traces in multiple surface receivers, but also computes the Bayesian evidence or the marginal likelihood that permits hypothesis testing for discriminating true vs. false event detection.