Abstract
Complex inference tasks, such as those encountered in Pulsar Timing Array(PTA) data analysis, rely on Bayesian frameworks. The high-dimensionalparameter space and the strong interdependencies among astrophysical, pulsarnoise, and nuisance parameters introduce significant challenges for efficientlearning and robust inference. These challenges are emblematic of broaderissues in decision science, where model over-parameterization and priorsensitivity can compromise both computational tractability and the reliabilityof the results. We address these issues in the framework of hierarchicalBayesian modeling by introducing a reparameterization strategy. Our approachemploys Normalizing Flows (NFs) to decorrelate the parameters governinghierarchical priors from those of astrophysical interest. The use of NF-basedmappings provides both the flexibility to realize the reparametrization and thetractability to preserve proper probability densities. We further adopti-nessai, a flow-guided nested sampler, to accelerate exploration of complexposteriors. This unified use of NFs improves statistical robustness andcomputational efficiency, providing a principled methodology for addressinghierarchical Bayesian inference in PTA analysis.