Computationally Recovering Preferred Factorizations of Quantum Hilbert Space

This thesis addresses the question of the preferred factorization of the quantum mechanical Hilbert Space into sub parts. Specifically, I computationally implement streamlined aspects of the recent “in principle” proposal of Carroll and Singh in their paper Quantum Mereology: Factorizing Hilbert Space into Subsystems with Quasi-Classical Dynamics. The goal is to link the selection of a preferred tensor product factorization to the appearance of quasi-classical behavior in this preferred factorization. Carroll and Singh quantify quasi-classical behavior through the criteria of “robustness” and “predictability” using the purity and pointer entropies. This work tests whether it is necessary to use both entropies or whether the minimization of purity entropy is sufficient in selecting quasi-classical behavior. This coding platform sets the stage for the application of machine learning to the problem of preferred basis