Approximating adiabatic transformation by alternating unitaries

Adiabatic transformation can be used in various quantum information processing, e.g., optimization, computation, search, etc. Recently, a gate-based algorithm which approximates adiabatic transformation by using alternating unitary operators was proposed [1]. In this talk, we introduce a concise version of this algorithm and conduct a benchmark test [2]. We find that approximation errors in the present algorithm cause broader distribution in the energy-eigenstate basis than a result of adiabatic control. As a result, it enables us to sample low energy eigenstates when it is applied to hard instances with exponentially small energy gaps.

[1] Dyon van Vreumingen, arXiv:2406.08064.

[2] Takuya Hatomura, arXiv:2407.12326.