\nQuantum interference between signal and background F eynman diagrams produce non-linear effects that challenge core assumpt ions going into the statistical analysis methodology in particle physi cs. I will show how for such cases\, no single observable can capture all the relevant information needed to perform optimal inference of th eory parameters from data collected in our experiments. The optimal da ta analysis strategy is to perform statistical inference directly on h igh-dimensional data\, without relying on summary histograms. Neural S imulation-Based Inference (NSBI) is a class of techniques that natural ly handle high dimensional data\, avoiding the need to design low-dime nsional summary histograms. We design a general purpose statistical fr amework in the ATLAS experiment that enables the application of NSBI t o full-scale physics analyses\, leading to the most precise measuremen t of the Higgs width by the experiment to date. \n \nThis work develop s several innovative solutions to introduce uncertainty quantification and enhance robustness and interpretability in NSBI. The developed me thod is an extension of the standard frequentist statistical inference framework used in particle physics and is therefore applicable to a w ide range of physics analyses. I will also discuss how this approach s implifies effective field theory interpretations in the high-dimension al space of theory parameters and future prospects. \n \n
\nBio: \ nDr. Aishik Ghosh is a postdoctoral scholar at UC Irvine and an affili ate at Berkeley Lab with a focus on Higgs physics at the ATLAS experim ent using novel statistical analysis methods and uncertainty quantific ation tools. His current efforts focus on the Higgs width and Higgs se lf-coupling measurements\, trigger algorithms and he also developed th e first generation of deep generative models for fast simulation of th e ATLAS calorimeter in 2018. Previously\, he obtained his PhD in parti cle physics from the University of Paris-Saclay also on the ATLAS expe riment. \n URL:https://www.physics.wisc.edu/events/?id=9098 END:VEVENT END:VCALENDAR