MSc. Niklas Götz

In this talk, I will explore the application of generative models, specifically neural importance sampling (NIS), for efficiently computing phase space integrals of cross sections in high-energy physics (HEP). Traditional Monte Carlo methods, such as the VEGAS algorithm, face challenges in accurately handling complex calculations required for modern experiments like those conducted at the LHC. This work focuses on bridging this gap by introducing ZüNIS, a fully automated NIS library tailored for HEP applications. I discuss extensions to the NIS formulation to improve stability and performance, along with the user-friendly design of ZüNIS, making it accessible to non-experts. Benchmark results on both toy and physics examples demonstrate both the effectiveness of ZüNIS, as well as the limitations of black box apporaches. The talk highlights the potential of generative models to revolutionize phase space integrals computation in HEP, paving the way for more efficient simulations in future experiments.