Authors: Jakab Nádori, Gregory Morse, Zita Majnay-Takács, Zoltán Zimborás, Péter Rakyta

Published on: February 07, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.05227

Summary

• What is new: A novel optimization method that navigates around the barren plateau (BP) issue in quantum circuit training by utilizing distant features of the cost-function landscape.
• Why this is important: The training of parameterized quantum circuits is hindered by the barren plateau problem, where gradient components exponentially diminish, making optimization difficult.
• What the research proposes: An optimization method that selects the search direction based on distant features of the cost-function landscape to effectively navigate around barren plateaus.
• Results: Applied to quantum circuits with 16 qubits and 15000 entangling gates, the method demonstrated resistance against BPs. An evolutionary selection framework further improved efficiency, outperforming traditional gradient-based approaches in quantum gate synthesis.

Technical Details

Technological frameworks used: Evolutionary selection framework

Models used: Variational quantum algorithms

Data used: Quantum circuits comprising 16 qubits and 15000 entangling gates

Potential Impact

This research could disrupt quantum computing companies and markets, particularly those focusing on quantum algorithm development, quantum circuit design, and optimization technologies.

Want to implement this idea in a business?

We have generated a startup concept here: QuantumPath.