TL;DR
This paper introduces a new computational tool, the "principal Landau determinant" (PLD), to precisely identify the singularities of Feynman integrals, which are crucial for high-precision calculations in particle physics. By adapting advanced algebraic geometry techniques, the authors provide a practical algorithm and software (PLD.jl) that can tackle complex diagrams previously intractable.
What Problem Does This Paper Address?
In particle physics, calculating the behavior of fundamental particles requires evaluating complex mathematical objects called Feynman integrals. A critical challenge is understanding where these integrals become singular (i.e., undefined), which often occurs at specific values of kinematic variables. These singularities, known as Landau singularities, are notoriously difficult to pinpoint, especially for complex diagrams involving massless particles or divergences. Existing methods have struggled with practical application to realistic Standard Model calculations, lacking systematic ways to classify and compute these singularities. This paper aims to bridge that gap by providing a rigorous and computationally feasible approach.
Methods
The core of this work is the introduction of the "principal Landau determinant" (PLD), a novel algebraic object derived from the GKZ (Gelfand-Kapranov-Zelevinsky) framework for generalized Euler integrals. The authors adapt these powerful tools, which are rooted in polyhedral geometry and symbolic/numerical elimination, to the specific context of Feynman diagrams. They define the PLD as a variety that captures components of the Landau singular locus. The paper details two primary computational approaches:
- Symbolic Elimination: This method leverages computer algebra systems to precisely compute the PLD by solving systems of polynomial equations. It's exact but can be computationally intensive for large diagrams.
- Numerical Elimination: For more complex cases, the authors developed a numerical algorithm that uses sampling and interpolation. This approach, implemented in their Julia package PLD.jl, provides a practical way to find approximate PLDs, which can then be rationalized to exact coefficients.
Datasets
The paper introduces a database of 114 Feynman diagrams, including various topologies and mass configurations, which are used to test and illustrate their algorithms. This database is made publicly available.
Key Insights and Ideas
- Bridging Theory and Practice: The PLD offers a more computationally tractable way to identify Landau singularities compared to traditional methods, making it applicable to realistic particle physics calculations.
- GKZ Framework Adaptation: The success of the GKZ system for generalized Euler integrals is adapted to Feynman integrals, demonstrating the power of these advanced algebraic tools in a new domain.
- Importance of "Dominant Components": The authors highlight that simply discarding "dominant components" (often associated with UV/IR divergences) from intermediate calculations can lead to missing crucial singularities, a pitfall their method avoids.
- Software Implementation: The development of the open-source Julia package PLD.jl is a significant contribution, providing a practical tool for researchers to compute Landau singularities.
Key Results
- The paper defines the principal Landau determinant (PLD) and provides algorithms for its computation, both symbolically and numerically.
- They demonstrate the effectiveness of their approach on a range of examples, including one-loop and banana diagrams, as well as cutting-edge Standard Model processes.
- The PLD.jl package is released, offering a user-friendly interface to their algorithms and including a database of results for 114 diagrams.
- Comparisons with existing tools like HyperInt show that PLD.jl can identify singularities that other methods might miss or struggle with.
Limitations and Future Work
- Computational Complexity: While the PLD offers an improvement, symbolic computation can still be intractable for very large Feynman diagrams.
- Numerical Interpolation Challenges: The numerical method relies on sampling and interpolation, which can be sensitive to the number of samples and may require careful parameter tuning for accuracy. The authors acknowledge that their current implementation might not handle "non-reduced components" perfectly.
- Conjectural Formulas: For certain cases, such as vanishing internal and external masses, the authors provide conjectural formulas for the PLD, which still require rigorous proof.
- Future Directions: The authors suggest exploring sparsity in PLD factors, investigating compactifications of Schwinger parameter space for improved results, and understanding the precise relationships between the Landau variety, Euler discriminant, and PLD. They also aim to improve the algorithms for handling more complex cases and to extend their work to include numerator factors.
Summary updated