R. D. Field Research Interests
CDF Public WEBsite The Collider Detector at Fermilab
Perturbative QCD
Collider Phenomenology
Neural Networks

Over the years most of my research has involved the extraction of measurable predictions from the theory of Quantum Chromodynamics (QCD). QCD is a precise and complete theory of quarks and gluons which we believe is the ultimate explanation of all strong interaction experiments at all energies, high and low. However, mathematical complexity makes it difficult to make precise predictions. The primary obstruction is the fact that the fundamental quarks and gluons of QCD apparently cannot be isolated as free particles, but are always confined within hadrons by strong forces not amenable to treatment by perturbative methods. Nevertheless, because QCD is an asymptotically free theory, interaction forces become weak at small distances (large energies) and calculations using perturbation theory and Feynman diagrams are possible. Unfortunately, most processes involve both low and high energy aspects, and it is difficult to separate the low energy pieces, which are not calculable by perturbative methods, from the high energy perturbative parts. The nonperturbative (low energy) pieces must be parameterized, taken from data, or a model is built to describe the regime.

Monte-Carlo techniques can be used to generate large transverse momentum events in hadron-hadron collisions that contain all the features expected from perturbative QCD. Incoming and outgoing partons are allowed to radiate (or Bremstrahlung) additional gluons. I find this an interesting area of research and I have used Monte-Carlo models to simulate high energy experimental events at CERN and Fermilab (Click to see collider events). In particluar, I have used events simulation techniques to explore the detection of the heavy Higgs boson at ultra high energy colliders. In a paper entitled, Enhancing the Heavy Higgs Signal with Jet-Jet Profile Cuts, Paul Griffin and I examined ZZ decay mode of the Higgs and in, Enhancing the Heavy Higgs to WW Signal at Hadron-Hadron Colliders, we extended our analysis to include the WW decay mode.

For several year my graduate students and I studied neural networks and other processing techniques as tools for high energy collider phenomenology. The great challenge at hadron colliders is to disentangle any new physics that may be present from the "ordinary" QCD background. Hadron collider events can be very complicated and quite often one has the situation where the signal is hiding beneath the background. In addition, there are many variables that describe a high energy collider event and it is not always obvious which variables best isolate the signal or precisely what data selection (or cuts) optimally enhance the signal over the background. Here neural networks are an excellent tool since they are ideal for separating patterns into categories (signal and background). We are able to "train" a networks to distinguish between signal and background using many variables to describe each event. The network computes a single variable that ranges from zero to one. When the training is successful the network will output a number near one for a signal event and near zero for a background event and a single cut can be made on the network output which will enhance the signal over the background.

In a paper entitled, Using Neural Networks to Enhance the Higgs Boson Signal at Hadron Colliders, we demonstrated that neural networks are a useful tool in Higgs boson phenomenology. In a paper entitled, A Topological Analysis of the Top Quark Signal and Background at Hadron Colliders, neural networks and Fisher discriminates are used in conjunction with modified Fox-Wolfram "shape" variables to help distinguish the top-pair signal from the W+jets and b-pair+jets background in proton-antiproton collisions at 1.8 TeV. In a paper entitled, Using Collider Event Topology in the Search for the Six-Jet Decay of Top Quark-Antiquark Pairs, the modified Fox-Wolfram "shape" variables are employed to help distinguish the top-pair signal from the ordinary QCD multi-jet background. Events are required to lie in a region of Hl-space determined by a genetic algorithm (GA) procedure to maximize the signal over the square root of the background.

I continue to be interested in making perturbative QCD calculations and comparing the results with experiment. In a paper entitled, Spin Dependent Drell-Yan in QCD to Order Alpha-Strong Squared: The Non-Singlet Sector, I collaborated with Sanghyeon Chang, Claudio Coriano, and L. E. Gordon to calculate the order alpha-strong squared corrections to the Drell-Yan (non-singlet) differential cross section for incoming states of arbitrary longitudinal helicities (i.e. polarized Drell-Yan.

On January 29, 1998, I joined the CDF collaboration at Fermilab and over the next several years the majority of my reseach will be involved with CDF and with CDF related phenomenology. The Collider Detector at Fermilab (CDF) is a general purpose experiment for the study of proton-antiproton collisions at a center-of-mass energy of 1.8 TeV at the Fermilab Tevatron Collider near Chicago, Illinois. For the next decade, the Fermilab Tevatron Collider remains the high energy frontier of particle physics. Precision capability at the energy frontier will allow will allow simultaneous attack on the open questions of high energy physics from many complementary directions including:

Each of these topics has the potential for revealing new physics and taken together they offer the most comprehensive discovery potential anywhere in particle physics for at least another decade. The CDF collaboration has begun taking data at a higher luminosity (called Run II). The detector was upgraded to handle this higher luminosity and is called CDF II. CDF II is a general purpose solenoidal detector which combines precision charged particle tracking with fast projective calorimetry and fined grained muon detection.

Recently I have been involved in the following CDF phenomenology:

Last modified: January 1, 2002