Reinforcement Learning for Quantum Circuit Design: Using Matrix Representations

AAAI 2025 Workshop CoLoRAI Submission6 Authors

Keywords: Quantum Circuit Design, Reinforcement Learning, Markov Decision Process, Circuit Matrix Representation, Q-Learning, Deep Q Network

TL;DR: In this paper, we applied Q-learning and DQN algorithms to two Matrix Representation on the quantum circuit design task, and we demonstrated that all four methods successfully discovered the expected quantum circuits for 10 gates.

Abstract: Quantum computing promises advantages over classical computing. The manufacturing of quantum hardware in the Noisy Intermediate-scale Quantum era is in its infancy stage. A major challenge of quantum circuit design is automatic mapping from a logical quantum circuit into gates from a universal quantum gate set. In this paper, we utilize reinforcement learning methods for the quantum circuit design task. By leveraging the power of reinforcement learning, we aim to provide an automatic and scalable approach over traditional hand-crafted heuristic methods.

Submission Number: 6