首页>
外文期刊>Physics of atomic nuclei
>Methodology for the Application of Deep Neural Networks in Searches for New Physics at Colliders and Statistical Interpretation of Expected Results
【24h】
Methodology for the Application of Deep Neural Networks in Searches for New Physics at Colliders and Statistical Interpretation of Expected Results
展开▼
机译:Methodology for the Application of Deep Neural Networks in Searches for New Physics at Colliders and Statistical Interpretation of Expected Results
Abstract Methods for the application of neural networks in searches for new physics in collider experiments are described. Mechanisms of selecting observables, optimizing hyperparameters, and constructing cascades of deep neural networks are considered. Methods for a statistical analysis of neural-network discriminants in theta and CombinedLimit packages with allowance for various uncertainties are presented.
展开▼