James S. Gainer,Joseph Lykken,Konstantin T. Matchev,Stephen Mrenna,Myeonghun Park
Abstract:The increasing use of multivariate methods, and in particular the Matrix Element Method (MEM), represents a revolution in experimental particle physics. With continued exponential growth in computing capabilities, the use of sophisticated multivariate methods-- already common-- will soon become ubiquitous and ultimately almost compulsory. While the existence of sophisticated algorithms for disentangling signal and background might naively suggest a diminished role for theorists, the use of the MEM, with its inherent connection to the calculation of differential cross sections will benefit from collaboration between theorists and experimentalists. In this white paper, we will briefly describe the MEM and some of its recent uses, note some current issues and potential resolutions, and speculate about exciting future opportunities.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: How to use the Matrix Element Method (MEM) more effectively in experimental particle physics to distinguish signals from backgrounds, thereby improving the sensitivity and precision of physical processes. Specifically, the paper discusses the history, current status, and future development directions of MEM, aiming to explore how to use the ever - increasing computing power to further optimize the application of MEM to meet the challenges of future high - energy physics experiments.
### Main problem decomposition
1. **Limitations of existing methods**:
- Currently, multivariate analysis methods (such as the Boosted Decision Tree (BDT) and neural networks) are widely used in experimental particle physics, but they may not be able to fully utilize the advantages of theoretical calculations.
- MEM, as a powerful tool, can provide the optimal test statistic, but it faces some challenges in practical applications, such as limited detector resolution, high - order corrections, and information loss.
2. **Application and improvement of MEM**:
- The paper reviews the applications of MEM in past research, for example, in studying the properties of top quarks at the Tevatron, B - physics, and the discovery of the Higgs boson.
- It proposes some problems and potential solutions in the current application of MEM, such as how to handle invisible particles, the parameterization of detector response functions, etc.
3. **Future development directions**:
- With the improvement of computing power, the paper looks forward to the possible future development directions of MEM, including but not limited to:
- **NLO/Parton Shower**: Extend MEM to consider additional radiation and other NLO corrections.
- **Jet sub - structure**: Use other properties of jets (such as sub - structure) to increase the sensitivity to partons in hard processes.
- **Underlying events, hadronization, etc.**: Combine more information about hadronization, underlying events, etc. to calculate a more complete likelihood function.
- **Detector resolution**: Use separate transfer functions for each detector element and consider time - dependence and correlations.
### Formula explanation
The paper gives the formula for MEM to calculate event likelihood:
\[
P(p_{vis}^i | \alpha) = \frac{1}{\sigma(\alpha)} \sum_{k,l} \int dx_1 dx_2 f_k(x_1) f_l(x_2) \frac{1}{2s x_1 x_2} \left[ \prod_{j \in inv.} \int \frac{d^3 p_j}{(2\pi)^3 2E_j} \right] |M_{kl}(p_{vis}^i, p_j; \alpha)|^2
\]
where:
- \( P(p_{vis}^i | \alpha) \) is the probability of observing momentum \( p_{vis}^i \) given the model parameter \(\alpha\).
- \( f_k(x_1) \) and \( f_l(x_2) \) are parton distribution functions.
- \( M_{kl}(p_{vis}^i, p_j; \alpha) \) is the theoretical matrix element.
- \( \sigma(\alpha) \) is the total cross - section, taking into account cuts and efficiencies.
Through this formula, MEM can provide a theoretically optimal test statistic, thus helping experimental physicists to distinguish signals from backgrounds more accurately.
### Conclusion
The paper emphasizes that with the continuous improvement of computing power, MEM will play a more important role in future high - energy physics experiments. However, this requires close cooperation between theoretical physicists and experimental physicists to overcome existing technical and computational challenges.