RESUMO
Graph neural networks have been shown to achieve excellent performance for several crucial tasks in particle physics, such as charged particle tracking, jet tagging, and clustering. An important domain for the application of these networks is the FGPA-based first layer of real-time data filtering at the CERN Large Hadron Collider, which has strict latency and resource constraints. We discuss how to design distance-weighted graph networks that can be executed with a latency of less than one µs on an FPGA. To do so, we consider a representative task associated to particle reconstruction and identification in a next-generation calorimeter operating at a particle collider. We use a graph network architecture developed for such purposes, and apply additional simplifications to match the computing constraints of Level-1 trigger systems, including weight quantization. Using the hls4ml library, we convert the compressed models into firmware to be implemented on an FPGA. Performance of the synthesized models is presented both in terms of inference accuracy and resource usage.
RESUMO
We present a method to establish, experimentally, the relation between the top-quark mass m_{t}^{MC} as implemented in Monte Carlo generators and the Lagrangian mass parameter m_{t} in a theoretically well-defined renormalization scheme. We propose a simultaneous fit of m_{t}^{MC} and an observable sensitive to m_{t}, which does not rely on any prior assumptions about the relation between m_{t} and m_{t}^{MC}. The measured observable is independent of m_{t}^{MC} and can be used subsequently for a determination of m_{t}. The analysis strategy is illustrated with examples for the extraction of m_{t} from inclusive and differential cross sections for hadroproduction of top quarks.
