Stanford Framework Improves Qubit Gate Count

Researchers at Stanford University have developed a new framework for optimizing superconducting quantum circuits, addressing a significant challenge in calculating gradients of eigenvalues and eigenvectors for complex Hamiltonians. The work uses the SQcircuit software package to enable efficient and precise computation of gradients, streamlining the design process for these circuits. Superconducting quantum circuits are uniquely suited to this gradient-based optimization due to their substantial flexibility in Hamiltonian design. Applying this framework to qubit discovery, the team demonstrated improvements in gate count, gate speed, decoherence time and resilience to noise compared to existing qubits.

Gradient-Based Optimization Framework for Superconducting Circuits

Performing an exhaustive search over a class of simple “single-loop” circuits with realistic parameter values, the team found fluxonium circuits, comprising a superinductor and single Josephson junction, performed best among the circuit types considered. This optimization was achieved using a new framework for gradient-based optimization of superconducting quantum circuits, which uses the SQcircuit software package for efficient computation of gradients. The framework’s versatility extends beyond qubit discovery, offering a pathway to address other optimization challenges in superconducting quantum hardware design.