MicroAlgo Launches Multi-Objective Evolutionary Algorithm for Automated Quantum Circuit Design
MLGO•MicroAlgo proposed a multi-objective evolutionary algorithm that automatically designs quantum circuits using a universal library by optimizing across accuracy, gate count, width and depth. In tests on Quantum Fourier Transform and Grover's algorithm it generated both textbook and novel circuit configurations within resource limits.
1. Innovation and Algorithm Overview
MicroAlgo’s new multi-objective evolutionary algorithm (MOEA) generates quantum circuits from scratch by sampling a universal library of components. It uses iterative crossover, mutation and selection to optimize multiple metrics—accuracy, gate count, circuit width and depth—simultaneously without predefined designs.
2. Validation on Quantum Fourier Transform and Grover’s Algorithm
The company applied the MOEA to design circuits for the Quantum Fourier Transform and Grover’s Search Algorithm. After multiple generations, the tool produced both standard textbook configurations and novel circuit structures that met input/output requirements under limited qubit and gate resources.
3. Technical Process and Industry Implications
The MOEA begins with random candidate circuits, simulates performance on accuracy and hardware metrics, then refines designs through genetic operations until optimal solutions emerge. This automation lowers barriers to quantum algorithm development and may accelerate practical applications on first-generation quantum processors.
