Ain Shams Engineering Journal

An approach for automated generation of quantum computing models using deep learning

Níyazí Furkan Bar, Mehmet Karaköse

Published Feb 28, 2025• 8 citations

Abstract

Quantum computing promises remarkable computational power with minimal energy consumption. However, the complexity of developing quantum circuits and codes hinders fully exploiting this potential. The study proposes an approach based on the automatic quantum circuit and code generation based on deep learning. It enables the resynthesis of existing circuits and the creation of new ones from undefined inputs. The system transforms inputs into reversible truth tables, generates a quantum unitary matrix, corrects errors, optimizes it, and converts it into a quantum code or circuit. This approach has been implemented on circuits and codes that involve up to five variables. Rigorous evaluations include both the Deep Neural Network and the overall approach. Although the DNN output does not guarantee absolute correctness, our approach compensates with supplementary processes, ensuring the precise generation of quantum codes and circuits. Comprehensive testing confirmed the approach's effectiveness in overcoming challenges in quantum circuit and code development. • Benefiting from the low energy and high computing power potential offered by quantum computing. • Developing quantum circuits and algorithms is a challenging task. • The number of competent people in the field who can develop quantum circuits and algorithms is low. • Classical algorithms do not work on quantum computers. • Insufficient number of quantum circuits and algorithms running on quantum computers.

DOI: 10.1016/j.asej.2025.103327

Related papers

##基于深度学习的量子电路优化

Jincheng Zhang

This paper explores the application of deep learning techniques to optimize quantum circuits. The core challenge in quantum computing lies in the intricate design and manipulation of quantum circuits to achieve desired computational outcomes. Traditional methods of quantum circuit optimization are often computationally expensive and require significant domain expertise. This research proposes a novel approach leveraging deep learning algorithms to automatically generate optimal quantum circuits. The system learns the underlying structure and characteristics of quantum circuits through training, effectively reducing the complexity associated with quantum computation. This work presents a framework for automated quantum circuit optimization, paving the way for more accessible and efficient quantum computing solutions. The key contribution is the integration of deep learning with quantum circuit optimization, offering an automated process for circuit design and potentially unlocking the full potential of quantum computers.

Sep 7, 20260 citations

Quantum Circuit Dataset Generator Approach for Deep Learning based Solutions

Níyazí Furkan Bar, Mehmet Karaköse

Today, although there are studies for the active use of quantum computers, deep learning-based studies on quantum computing are few. Since the large amount of data needed by deep learning cannot be met from the literature, using deep learning in subjects such as quantum circuit optimization, and quantum circuit/algorithm generation challenges researchers. In this study, an approach is proposed to overcome the data and dataset deficiencies related to quantum circuits in the literature. The proposed approach generates datasets consisting of quantum circuits, truth tables of quantum circuits, unitary matrices, visuals, and pairs of these data. The circuits, data, and datasets produced with this approach have already been studied and are still being actively done. The performance of the proposed approach was evaluated as dependent on different variables and as elapsed time. With the proposed approach, circuits, data, and datasets with desired properties can be generated successfully. As a result, the contribution and performance of the proposed approach have been verified.

Feb 21, 20246 citations

基于深度学习的量子态预测与优化

Jincheng Zhang

Quantum computing holds immense potential for revolutionizing various fields, but realizing this potential requires overcoming significant challenges related to qubit stability, coherence, and control. Quantum state prediction and optimization are crucial steps in this process. This paper introduces a novel deep learning-based approach for both quantum state prediction and optimization. We leverage the power of deep neural networks to learn the complex patterns within quantum states, enabling accurate predictions of quantum circuit performance and automated optimization of circuit design. The algorithm's effectiveness is demonstrated through theoretical analysis and preliminary experimental validation, showcasing a significant improvement over existing methods. This work represents a significant advancement in quantum algorithm design, offering a promising pathway towards more efficient and reliable quantum computations.

Sep 2, 20260 citations

Generating Quantum Software from Truth Tables

Macario Polo Usaola, Ignacio García‐Rodríguez de Guzmán, Manuel Serrano +1 more

Quantum computing promises significant advancements over classical computing by enabling exponential speedups for specific problems. Despite its potential, quantum software development remains challenging due to its complexity, cost, and susceptibility to errors. Aligned with the novel paradigm of Quantum Software Engineering (QSE), this paper introduces a tool designed to automate the generation of quantum software components from truth tables. The tool simplifies the development process by allowing users to specify expected output values for given inputs, which the tool then translates into the corresponding quantum code. This approach reduces the manual effort required and helps ensure accuracy in quantum program development. Our method not only streamlines quantum software generation but also optimizes the resulting quantum circuits by minimizing unnecessary computations. The paper provides a detailed overview of the algorithm, tool implementation, and potential applications, highlighting the tool's efficacy in facilitating more efficient quantum software development.

IEEE International Conference on Quantum Computing and Engineering (QCE)Sep 15, 20242 citations

On the Challenges of Quantum Circuit Encoding Using Deep and Reinforcement Learning

Patrick Selig, Niall Murphy, David Redmond +1 more

Quantum computers may solve some computing tasks more efficiently than classical computers, but to do so requires the design of appropriate quantum circuits. However, it is hard to design even simple quantum programs. Recent results have shown that deep learning augmented search has the potential to discover good circuits for a problem, but it is not yet sufficiently understood how well deep learning is able to model such a complex domain: the search space and output space grow exponentially. This work explores the ability of neural networks to encode quantum circuits for tasks that require knowledge about the unitary representation of the circuit. To this end, we trained neural networks to directly learn to predict the unitary of a circuit and applied reinforcement learning to train neural networks to solve circuit based quantum state preparation. This work finds that encoding quantum circuits is quite difficult especially with regards to the amount of entanglement applied in both application domains. The use of intermediate states and structure in quantum circuits combined with a reasonable inductive bias, where applicable, can alleviate these problems to some extent. The use of intermediate states also greatly improves the results of Deep Reinforcement Learning for quantum state preparation. Based on these results, we discuss the next set of challenges to address if we are to design neural network-based approaches for the automatic generation of quantum circuits. Add sentence on how it improves things in the end.

IEEE AccessJan 1, 20250 citations

AutoQC: Automated Synthesis of Quantum Circuits Using Neural Network

Murakami Kentaro, Jianjun Zhao

While the ability to build quantum computers is improving dramatically, developing quantum algorithms is limited and relies on human insight and ingenuity. Although a number of quantum programming languages have been developed, it is challenging for software developers who are not familiar with quantum computing to learn and use these languages. It is, therefore, necessary to develop tools to support developing new quantum algorithms and programs automatically. This paper proposes AutoQC, an approach to automatically synthesizing quantum circuits using the neural network from input and output pairs. We consider a quantum circuit a sequence of quantum gates and synthesize a quantum circuit probabilistically by prioritizing with a neural network at each step. The experimental results highlight the ability of AutoQC to synthesize some essential quantum circuits at a lower cost.

arXiv:2210.02766Oct 6, 20221 citation