A recursively partitioned approach to architecture-aware ZX Polynomial synthesis and optimization

Understand

The synthesis of quantum circuits from phase gadgets in the ZX-calculus facilitates quantum circuit optimization.

• Our work provides an alternative formulation for the architecture-aware synthesis algorithm of PauliOpt by replacing the stochastic approach of PauliOpt with a heuristic based search and utilizes a divide and conquer method to synthesize an optimized circuit from a ZX polynomial.
• We provide a comparison of our algorithm with PauliOpt and other state-of-the-art optimization libraries.
• While we note poorer performance for highly structured circuits, as in the QAOA formulation for Max-Cut, we demonstrate a significant advantage for randomized circuits, which highlights the advantages of utilizing an architecture-aware methodology.