This is an old revision of the document!
Signals of physics beyond the standard model (BSM) can often be masked by standard model (SM) processes. In tandem with my continuing work on model building and model interpretation for the LHC, I have started to study contemporary methods for estimating SM backgrounds.
The theoretical challenge in estimating backgrounds is that there are two fundamentally different ways of predicting SM rates: inclusive vs. exclusive. Inclusive calculations describe the gross properties of SM backgrounds and are very accurate. Exclusive calculations describe the detailed properties of SM backgrounds but are less accurate. Ideally, one would use both inclusive and exclusive information at the same time, but one has to be very careful not to double count, since exclusive predictions are supposed to be subsets of the inclusive prediction.
The has been a growing body of work in creating merged inclusive/exclusive calculations, and this merging is essential to making detailed predictions for the LHC. With Christian Bauer and Frank Tackmann, we developed a new strategy for crafting inclusive/exclusive results. We are writing a computer program called GenEvA (GENerate EVents Analytically) to implement this strategy. The name GenEvA is obviously inspired by the site of the upcoming LHC experiment.
Spring 2008 - Present
In progress…
Spring 2008 - Present
In progress…
Summer 2007 - Spring 2008
In principle, one can make predictions for SM backgrounds using pencil-and-paper alone. However, the complexity of contemporary calculations requires the use of computer tools. Also, in order to understand contemporary particle experiments one needs a computer model of the detector. Therefore, the easiest way to predict what particle experiment will see is to run pseudo-experiments on a computer (actually on thousands of computers). Because particle physics is based on quantum mechanics, and because quantum physics is inherently probabilistic, such computer simulations are referred to as Monte Carlo programs.
There are already many Monte Carlo programs available for making theoretical predictions, but these tools are divided into tools that can make inclusive predictions and those that can make exclusive predictions. As mentioned above, ideally one would make combined inclusive/exclusive predictions, but this is challenging for various reasons. On the theoretical side, the formalism for making combined predictions is in its infancy, and there are various competing proposals for how best to merge inclusive and exclusive calculations. On the algorithmic side, one needs a computer program that can handle hybrid calculations, and while there are concrete examples for how to do this, they usually rely on a separate engine for the inclusive piece and the exclusive piece.
In early 2008, we released a prototype computer program GenEvA (GENerate EVents Analytically) that is simultaneously a theoretical framework for how to build an inclusive/exclusive calculation as well as an algorithm for implementing that calculation. The idea is to start with an algorithm defined by the exclusive calculation and use that algorithm to implement the inclusive calculation. By construction, the resulting predictions will have both inclusive and exclusive information.
Because of the theoretical and technical hurdles for building such a program, the prototype version only handles electron-positron collisions and not the more complicated proton-proton collisions at the LHC. We hope to build an alpha version for proton-proton collisions by early 2009.