Quantum 2025 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
THU: Thursday Contributed Sessions
THU 13: Poster Session: Applications
THU 13.23: Poster
Thursday, September 11, 2025, 16:30–18:30, ZHG Foyer 1. OG
Optimal Parameters for Quantum Approximate Optimization via Weak Measurements — •Lena Wagner1,2 and Tobias Stollenwerk1 — 1Forschungszentrum Jülich — 2Universität des Saarlandes
Quantum imaginary time evolution (QITE) has emerged as a potentially promising method for ground state preparation in quantum computing, leveraging its inherent ability to amplify ground state amplitudes and exponentially suppress those of excited states. In this work, we focus on an QITE inspired approach based on weak measurements. Conditional unitary operations ("scrambling") are used to correct for the non-unitary evolution and undesired measurement outcomes. These scrambling operations are implemented via rotations introducing optimizable angles and activate only when measurement outcomes exceed a certain threshold. Further tunable parameters include a scaling factor for energy spectrum normalization and the choice of an initial state with non-zero overlap with the ground state. By generalizing this algorithm from exact to approximate optimization, we enable applications to combinatorial optimization problems such as MaxCut. We investigate performance of the algorithm with respect to these parameters and present parameter studies on MaxCut instances, revealing how scrambling angles, threshold, scaling factors, and the randomness of the weak measurement impact convergence and solution quality.
Keywords: Quantum Computing Algorithms; Quantum Optimization; Quantum Simulation