Search

Abstract: We discuss several popular statistical learning methods used in high-energy- and astro-physics analysis. After a short motivation for statistical learning we present the most popular algorithms and discuss several examples from current research in particle- and astro-physics. The statistical learning methods are compared with each other and with standard methods for the respective application. [Copyright &y& Elsevier]

Abstract: For LHC, the hadronic calorimetry of the ATLAS detector is performed by Tilecal, a scintillating tile calorimeter. For calibration purposes, a fraction of the Tilecal modules is placed in particle beam lines. Despite beam high quality, experimental beam contamination is observed and this masks the actual performance of the calorimeter. For optimizing the calibration task, an online neural particle classifier was developed for Tilecal. Envisaging a neural trigger for incoming particles, a neural process runs integrated to the data acquisition task and performs online training for particle identification. The neural classification performance is evaluated by correlating the neural response to classical methodology, confirming an ability for outsider identification at levels as high as 99.3%. [Copyright &y& Elsevier]

Abstract: The paper is devoted to the problem of time series enhancement, which is based on the analysis of local regularity. The model construction using this analysis does not require any a priori assumption on the structure of the noise and the functional relationship between original signal and noise. The signal itself may be nowhere differentiable with rapidly varying local regularity, what is overcome with the help of the new technique of increasing the local Hölder regularity of the signal under research. A new signal with prescribed regularity is constructed using the genetic algorithm. This approach is applied to enhancement of time series in the paleoclimatology, solar physics, dendrochronology, meteorology and hydrology. [Copyright &y& Elsevier]