Authors: Yash J. Patel, Akash Kundu, Mateusz Ostaszewski, Xavier Bonet-Monroig, Vedran Dunjko, Onur Danaci

Published on: February 05, 2024

Impact Score: 8.22

Arxiv code: Arxiv:2402.03500

Summary

• What is new: Introduction of a curriculum-based reinforcement learning quantum architecture search (CRLQAS) algorithm designed for the noisy intermediate-scale quantum era, focusing on overcoming limitations in circuit design due to noise.
• Why this is important: Difficulty in finding useful quantum circuit architectures compatible with current device limitations, and the significant impact of noise on both parameter optimization and architecture search performance.
• What the research proposes: A CRLQAS algorithm that effectively searches for optimal quantum circuit architectures by incorporating a unique 3D architecture encoding, an episode halting scheme for shorter circuits, and a novel optimization technique.
• Results: CRLQAS demonstrates superior performance in designing circuits for quantum chemistry tasks, outperforming existing algorithms in both noiseless and noisy conditions.

Technical Details

Technological frameworks used: Curriculum-based reinforcement learning, Pauli-transfer matrix formalism

Models used: Simultaneous Perturbation Stochastic Approximation variant as an optimizer

Data used: Quantum chemistry tasks

Potential Impact

Quantum computing, particularly companies focusing on quantum chemistry and quantum algorithm development.

Want to implement this idea in a business?

We have generated a startup concept here: QuantOptima.