ABSTRACT
This chapter starts with an introduction to the models most relevant for Monte Carlo simulations, followed by a selection of applications. It reviews strong and weak points about the utilization of Neural Networks applications for Monte Carlo simulations. Neural Networks are a specific branch of the Artificial Intelligence domain in computer science. They get their inspiration from the fact that humans are evidently able to fulfill complex tasks; hence, by replicating the low-level mechanisms of the human brain on computing systems, one can potentially construct high level algorithms with similar capabilities. The typical utilization pattern for a majority of network topologies is to feed them as input a large set of data representing the problem of interest, be it a medical image, a set of features or any output from the instrumentation, and at the same time provide the “expected output” from a so-called training set.
