Trigger Simulations for GRB Detection with the Swift Burst Alert Telescope

Lien, A. Y.; Sakamoto, T.; Gehrels, N.; Palmer, D. M.; Graziani, C.

doi:10.1017/s1743921312013282

Cited by 1 publication

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To investigate this issue further, a recent study by Lien et al (2012) performed a Monte Carlo analysis that generated a mock sample of GRBs, using the GRB rate and luminosity function of Wanderman & Piran (2010), and processed them through an entire simulated Swift detection pipeline, applying the full set of Swift trigger criteria, to determine which GRBs would be detected. It was found that the resulting measured GRB rate as a function of redshift followed very closely that of the true Swift GRB set described in Fynbo et al (2009); this finding is consistent with both the mock GRB sample and the simulated trigger pipeline being good approximations to reality.…”

Section: Classification: Identifying Gamma-ray Burstersmentioning

confidence: 99%

“…Our goal here is to replace the simulated Swift trigger pipeline with a classification NN, which (as we will see) can determine in just a few microseconds whether a given GRB is detected. To this end, we use as training data a pre-computed mock sample of 10, 000 GRBs from Lien et al (2012). In particular, we divide this sample randomly into ∼ 4000 for training, ∼ 1000 for validation, and the final ∼ 5000 as a blind test set on which to perform our final evaluations.…”

Section: Form Of the Classification Problemmentioning

confidence: 99%

See 1 more Smart Citation

SkyNet: an efficient and robust neural network training tool for machine learning in astronomy

Graff

Feroz²,

Hobson³

et al. 2014

Monthly Notices of the Royal Astronomical Society

110

View full text Add to dashboard Cite

We present the first public release of our generic neural network training algorithm, called SKYNET. This efficient and robust machine learning tool is able to train large and deep feedforward neural networks, including autoencoders, for use in a wide range of supervised and unsupervised learning applications, such as regression, classification, density estimation, clustering and dimensionality reduction. SKYNET uses a 'pre-training' method to obtain a set of network parameters that has empirically been shown to be close to a good solution, followed by further optimisation using a regularised variant of Newton's method, where the level of regularisation is determined and adjusted automatically; the latter uses second-order derivative information to improve convergence, but without the need to evaluate or store the full Hessian matrix, by using a fast approximate method to calculate Hessian-vector products. This combination of methods allows for the training of complicated networks that are difficult to optimise using standard backpropagation techniques. SKYNET employs convergence criteria that naturally prevent overfitting, and also includes a fast algorithm for estimating the accuracy of network outputs. The utility and flexibility of SKYNET are demonstrated by application to a number of toy problems, and to astronomical problems focusing on the recovery of structure from blurred and noisy images, the identification of gammaray bursters, and the compression and denoising of galaxy images. The SKYNET software, which is implemented in standard ANSI C and fully parallelised using MPI, is available at

show abstract

Section: Classification: Identifying Gamma-ray Burstersmentioning

confidence: 99%

Section: Form Of the Classification Problemmentioning

confidence: 99%