GANplifying event samples
Anja Butter, Sascha Diefenbacher, Gregor Kasieczka, Benjamin Nachman, Tilman Plehn
Abstract
A critical question concerning generative networks applied to event generation in particle physics is if the generated events add statistical precision beyond the training sample. We show for a simple example with increasing dimensionality how generative networks indeed amplify the training statistics. We quantify their impact through an amplification factor or equivalent numbers of sampled events.
Topics & Concepts
Event (particle physics)Computer scienceCurse of dimensionalitySimple (philosophy)Generative grammarArtificial intelligenceSample (material)Generative modelStatistical modelStatistical analysisArtificial neural networkTraining (meteorology)Dimension (graph theory)Machine learningStatistical hypothesis testingMathematicsPattern recognition (psychology)Feature (linguistics)StatisticsTraining setData miningComponent (thermodynamics)Quantum Mechanics and ApplicationsQuantum many-body systemsParticle physics theoretical and experimental studies