Detecting quantum phase transitions via shallow variational quantum circuits

Research Question and Approach

This paper investigates the reliability and underlying mechanisms of Delta-VQE, a hybrid quantum-classical algorithm designed to detect quantum phase transitions. Unlike traditional methods that require precise ground-state preparation, Delta-VQE identifies critical points by comparing the variational energies of two trial states, each initialized to represent a different phase of matter. The authors test this approach on a one-dimensional transverse-field Ising model with a three-spin cluster interaction, a system that exhibits both true Ising critical points and a distinct self-dual line.

The study reveals that the effectiveness of Delta-VQE is not universal but depends heavily on the construction of the variational ansätze. When the researchers employed dual ansätze—circuits that map onto each other via the system's duality transformation—the algorithm consistently identified the self-dual line (h=1) as the critical point, even though the true phase transitions occur elsewhere. However, when the ansätze were specifically tailored to embody the physics of the competing phases, the algorithm successfully pinpointed the genuine Ising critical points. The authors demonstrate that while the method is resource-efficient and works well with shallow circuits, its diagnostic output is a direct consequence of the physical representativeness of the chosen initial states.

As quantum hardware remains in the noisy intermediate-scale quantum (NISQ) regime, resource-efficient algorithms like Delta-VQE are essential for exploring many-body physics. This work provides a crucial cautionary note for researchers: the simplicity of the Delta-VQE diagnostic can be misleading. It highlights that without a careful, physics-informed selection of trial states, the algorithm may produce "false positives" by detecting mathematical symmetries (like self-duality) rather than physical phase transitions. This underscores the necessity of integrating domain knowledge into quantum circuit design.