Automation;
Higher-Order Perturbative Calculations;
Higher Order Electroweak Calculations;
LOOP;
PARTS;
D O I:
10.1007/JHEP05(2022)161
中图分类号:
O412 [相对论、场论];
O572.2 [粒子物理学];
学科分类号:
摘要:
We present a new and fully general algorithm for the automated construction of the integrands of two-loop scattering amplitudes. This is achieved through a generalisation of the open-loops method to two loops. The core of the algorithm consists of a numerical recursion, where the various building blocks of two-loop diagrams are connected to each other through process-independent operations that depend only on the Feynman rules of the model at hand. This recursion is implemented in terms of tensor coefficients that encode the polynomial dependence of loop numerators on the two independent loop momenta. The resulting coefficients are ready to be combined with corresponding tensor integrals to form scattering probability densities at two loops. To optimise CPU efficiency we have compared several algorithmic options identifying one that outperforms naive solutions by two orders of magnitude. This new algorithm is implemented in the OpenLoops framework in a fully automated way for two-loop QED and QCD corrections to any Standard Model process. The technical performance is discussed in detail for several 2 -> 2 and 2 -> 3 processes with up to order 10(5) two-loop diagrams. We find that the CPU cost scales linearly with the number of two-loop diagrams and is comparable to the cost of corresponding real-virtual ingredients in a NNLO calculation. This new algorithm constitutes a key building block for the construction of an automated generator of scattering amplitudes at two loops.
机构:
Inst Modern Phys, NanChangLu 509, Lanzhou 730000, Peoples R ChinaInst Modern Phys, NanChangLu 509, Lanzhou 730000, Peoples R China
Evslin, Jarah
Guo, Hengyuan
论文数: 0引用数: 0
h-index: 0
机构:
Inst Modern Phys, NanChangLu 509, Lanzhou 730000, Peoples R China
Univ Chinese Acad Sci, YuQuanLu 19A, Beijing 100049, Peoples R ChinaInst Modern Phys, NanChangLu 509, Lanzhou 730000, Peoples R China