Springer Theses Development of Machine Learning τ Trigger Algorithms and Search for Higgs Boson Pair Production

This book reports the successful optimization of the Compact Mupn Solenoid (CMS) tau trigger algorithm for the Run-3 (Phase-1) of the Large Hadron Collider (LHC) and a completely new and original design of a machine learning based tau triggering algorithm for the High Luminosity LHC (or Phase-2). A large proportion of searches at collider experiments relies on datasets collected with a dedicated tau lepton selection algorithm, particularly difficult to operate in intense hadronic environments, making the work descirbed in this book of prime importance. The second part of the book describes a major and very challenging data analysis, aiming to detect Higgs boson pair production. The book summarizes these contributions in clear, pedagogical prose while keeping an adequate and coherent balance between the technical and data analysis aspects. Machine learning techniques were used extensively throughout this research; therefore, special care has been taken to describe their core principles and application in high-energy physics, as well as potential future developments for sophisticated low-latency trigger algorithms and modern signal extraction methods. This book reports the successful optimization of the Compact Mupn Solenoid (CMS) tau trigger algorithm for the Run-3 (Phase-1) of the Large Hadron Collider (LHC) and a completely new and original design of a machine learning based tau triggering algorithm for the High Luminosity LHC (or Phase-2). A large proportion of searches at collider experiments relies on datasets collected with a dedicated tau lepton selection algorithm, particularly difficult to operate in intense hadronic environments, making the work descirbed in this book of prime importance. The second part of the book describes a major and very challenging data analysis, aiming to detect Higgs boson pair production. The book summarizes these contributions in clear, pedagogical prose while keeping an adequate and coherent balance between the technical and data analysis aspects. Machine learning techniques were used extensively throughout this research; therefore, special care has been taken to describe their core principles and application in high-energy physics, as well as potential future developments for sophisticated low-latency trigger algorithms and modern signal extraction methods.