The use of adversaries for optimal neural network training

Abstract

B-decay data from the Belle experiment at the KEKB collider have a substantial background from $e^{+}{e}^{-}\to q\bar{q}$ events. To suppress this we employ deep neural network algorithms. These provide improved signal from background discrimination. However, the deep neural network develops a substantial correlation with the $\Delta E$ kinematic variable used to distinguish signal from background in the final fit due to its relationship with input variables. The effect of this correlation is reduced by deploying an adversarial neural network. Overall the adversarial deep neural network performs better than a Boosted Decision Tree algorithimn and a commercial package, NeuroBayes, which employs a neural net with a single hidden layer.

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Generative Adversarial Networks
• A Theory of Dark Matter
• Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization
• Learning to Pivot with Adversarial Networks
• A Neural Bayesian Estimator for Conditional Probability Densities
• Diagnostic for new physics in $B\to \pi K$ decays

Cited by in corpus (2)

• QCD or What?
• Quantum Support Vector Machines for Continuum Suppression in B Meson Decays