Sunil Jaiswal
The Ohio State University
Bayesian framework for model-data comparison incorporating theoretical uncertainties: Application to heavy-ion collisions
Location: 2128 Physics Research Building
Abstract: Comparison of theoretical models with experimental data requires careful consideration of both experimental and theoretical uncertainties. Deficiencies in theoretical models can lead to physical parameters varying significantly depending on the choice of theory and/or experimental data, highlighting the importance of properly accounting for theoretical uncertainties. In this seminar, I will present a Bayesian framework that explicitly quantifies these uncertainties by statistically modeling theoretical errors, guided by qualitative knowledge of the theory’s domain of reliability. The approach enables rigorous, uncertainty-quantified Bayesian model–data comparison. The concepts will be discussed using a simple system: a ball drop experiment. I will then present data-driven, uncertainty-aware constraints on the specific shear and bulk viscosities of the quark–gluon plasma.