Inference of Experimental Radial Impurity Transport on Alcator C-Mod: Bayesian Parameter Estimation and Model Selection

Sciortino, F.; Howard, N. T.; Marmar, E. S.; Odstrcil, T.; Cao, N. M.; Dux, R.; Hubbard, A. E.; Hughes, J. W.; Irby, J. H.; Marzouk, Y.; Milanese, L. M.; Reinke, M. L.; Rice, J. E.; Rodriguez-Fernandez, P.

doi:10.48550/arxiv.2006.06798

Cited by 1 publication

(1 citation statement)

References 87 publications

(135 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, these methods are only for finding the lenses, and a mass model is necessary for further studies. Mass models of gravitational lenses are often described by parametrized profiles, where the parameters are optimized for instance via Markov-Chain Monte-Carlo (MCMC) sampling (e.g., Jullo et al 2007;Suyu & Halkola 2010;Sciortino et al 2020;Fowlie et al 2020). Such techniques are very time and resource consuming and are thus difficult to scale up for the upcoming amount of data.…”

Section: Introductionmentioning

confidence: 99%

HOLISMOKES -- IV. Efficient mass modeling of strong lenses through deep learning

Schuldt,

Suyu,

Meinhardt

et al. 2020

Preprint

View full text Add to dashboard Cite

Modelling the mass distributions of strong gravitational lenses is often necessary to use them as astrophysical and cosmological probes. With the high number of lens systems ( 10 5 ) expected from upcoming surveys, it is timely to explore efficient modeling approaches beyond traditional Markov-Chain Monte-Carlo techniques that are time consuming. We train a Convolutional Neural Network (CNN) on images of galaxy-scale lens systems to predict the five parameters of the Singular Isothermal Ellipsoid (SIE) mass model (lens center x and y, complex ellipticity e x and e y , and Einstein radius θ E ). To train the network, we simulate images based on real observations from the Hyper Suprime-Cam Survey for the lens galaxies and from the Hubble Ultra Deep Field as lensed galaxies. We tested different network architectures and the effect of different data sets such as using only double or quad systems defined based on the source center, and using different input distributions of θ E . We find that the CNN performs well and obtain with the network trained on both doubles and quads with a uniform distribution of θ E > 0.5 the following median values with 1σ scatter: ∆x = (0.00 +0.30 −0.30 ) , ∆y = (0.00 +0.30 −0.29 ) , ∆θ E = (0.07 +0.29 −0.12 ) , ∆e x = −0.01 +0.08 −0.09 and ∆e y = 0.00 +0.08 −0.09 . The bias in θ E is driven by systems with small θ E . Therefore, when we further predict the multiple lensed image positions and time delays based on the network output, we apply the network to the sample limited to θ E > 0.8 . In this case, the offset between the predicted and input lensed image positions is (0.00 +0.29 −0.29 ) and (0.00 +0.32 −0.31 ) for the x and y coordinates, respectively. For the fractional difference between the predicted and true time delay, we obtain 0.04 +0.27 −0.05 . Our CNN model is able to predict the SIE parameter values in fractions of a second on a single CPU and with the output we can predict the image positions and time delays in an automated way, such that we are able to process efficiently the huge amount of expected galaxy-scale lens detections in the near future.

show abstract

Section: Introductionmentioning

confidence: 99%