Analysing Particle Jets with Artificial Neural Networks

Elementary particle physics includes a number of feature recognition problems for which artificial neural networks can be used. We used a feed-forward neural network to seperate particle jets originating from b-quarks from other jets. Some aspects such as pruning and overfitting have been studied. Furthermore, the influence of modifications in architecture and input space have been examined. In addtition we discuss how self-organizing networks can be applied to high energy physics problems.