Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userOptimization of Signal Significance by Bagging Decision Trees
88
0
0.0
(
0
)
Authors
I. Narsky
Ask ChatGPT about the research
No Arabic abstract
An algorithm for optimization of signal significance or any other classification figure of merit suited for analysis of high energy physics (HEP) data is described. This algorithm trains decision trees on many bootstrap replicas of training data with each tree required to optimize the signal significance or any other chosen figure of merit. New data are then classified by a simple majority vote of the built trees. The performance of this algorithm has been studied using a search for the radiative leptonic decay B->gamma l nu at BaBar and shown to be superior to that of all other attempted classifiers including such powerful methods as boosted decision trees. In the B->gamma e nu channel, the described algorithm increases the expected signal significance from 2.4 sigma obtained by an original method designed for the B->gamma l nu analysis to 3.0 sigma.
rate research
Read More
The experimental issue of the search for new particles of unknown mass poses the challenge of exploring a wide interval to look for the usual signatures represented by excess of events above the background. A side effect of such a broad range quest is that the traditional significance calculations valid for signals of known location are no more applicable when such an information is missing. In this note the specific signal search approach via observation windows sliding over the range of interest is considered; in the assumptions of known background and of fixed width of the exploring windows the statistical implications of such a search scheme are described, with special emphasis on the correct significance assessment for a claimed discovery.
In this paper we recreate, and improve, the binary classification method for particles proposed in Roe et al. (2005) paper Boosted decision trees as an alternative to artificial neural networks for particle identification. Such particles are tau neutrinos, which we will refer to as background, and electronic neutrinos: the signal we are interested in. In the original paper the preferred algorithm is a Boosted decision tree. This is due to its low effort tuning and good overall performance at the time. Our choice for implementation is a deep neural network, faster and more promising in performance. We will show how, using modern techniques, we are able to improve on the original result, both in accuracy and in training time.
We discuss the traditional criterion for discovery in Particle Physics of requiring a significance corresponding to at least 5 sigma; and whether a more nuanced approach might be better.
Particle may sometimes have energy outside the range of radiation detection hardware so that the signal is saturated and useful information is lost. We have therefore investigated the possibility of using an Artificial Neural Network (ANN) to restore the saturated waveforms of $gamma$ signals. Several ANNs were tested, namely the Back Propagation (BP), Simple Recurrent (Elman), Radical Basis Function (RBF) and Generalized Radial Basis Function (GRBF) neural networks (NNs) and compared with the fitting method based on the Marrone model. The GBRFNN was found to perform best.
Signal estimation in the presence of background noise is a common problem in several scientific disciplines. An On/Off measurement is performed when the background itself is not known, being estimated from a background control sample. The frequentist and Bayesian approaches for signal estimation in On/Off measurements are reviewed and compared, focusing on the weakness of the former and on the advantages of the latter in correctly addressing the Poissonian nature of the problem. In this work, we devise a novel reconstruction method, dubbed BASiL (Bayesian Analysis including Single-event Likelihoods), for estimating the signal rate based on the Bayesian formalism. It uses information on event-by-event individual parameters and their distribution for the signal and background population. Events are thereby weighted according to their likelihood of being a signal or a background event and background suppression can be achieved without performing fixed fiducial cuts. Throughout the work, we maintain a general notation, that allows to apply the method generically, and provide a performance test using real data and simulations of observations with the MAGIC telescopes, as demonstration of the performance for Cherenkov telescopes. BASiL allows to estimate the signal more precisely, avoiding loss of exposure due to signal extraction cuts. We expect its applicability to be straightforward in similar cases.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
