Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Information Communication Technologies Conference (ICTC)

Meta-Learning with QAOA for Channel Decoding

Jun Zhu, Xutao Yu, Han Zeng

Published May 10, 2024• 1 citation
Publisher

Abstract

With the development of 6G networks, the required computational power has been greatly increased, and quantum computing offers promising approaches for this purpose due to its properties such as superposition and entanglement. It has been proposed to combine the Quantum Approximation optimization Algorithm (QAOA) with the Learning-to-Learn (L2L) strategy using classical Recurrent Neural Networks (RNN) to solve the problem of channel decoding on Noisy IntermediateScale Quantum (NISQ) devices. In this paper, we build on this foundation to further investigate the impact of different RNNs on channel decoding. The experimental results show that by training the RNN model of Minimal Gated Unit (MGU) with fewer parameters, the initial parameters of the obtained QAOA have a certain degree of similarity with the Long Short Term Memory Network (LSTM), and the decoding results are comparable. The method of training RNNs improves the performance and saves the cost compared to the random initialization method. Despite the limitation by the current quantum hardware, this quantum computation approach provides the future for channel decoding in the 6G network a promising scheme.

DOI
10.1109/ictc61510.2024.10602144

Related papers

Learning to Learn With Quantum Optimization Via Quantum Neural Networks

Kuan‐Cheng Chen, Hiromichi Matsuyama, Wei‐Hao Huang

Quantum Approximate Optimization Algorithms (QAOA) promise efficient solutions to classically intractable combinatorial optimization problems by harnessing shallowdepth quantum circuits. Yet, their performance and scalability often hinge on effective parameter optimization, which remains nontrivial due to rugged energy landscapes and hardware noise. In this work, we introduce a quantum meta-learning framework that combines quantum neural networks, specifically Quantum Long Short-Term Memory (QLSTM) architectures, with QAOA. By training the QLSTM optimizer on smaller graph instances, our approach rapidly generalizes to larger, more complex problems, substantially reducing the number of iterations required for convergence. Through comprehensive benchmarks on Max-Cut and Sherrington-Kirkpatrick model instances, we demonstrate that QLSTM-based optimizers converge faster and achieve higher approximation ratios compared to classical baselines, thereby offering a robust pathway toward scalable quantum optimization in the NISQ era.

IEEE International Conference on Quantum Computing and Engineering (QCE)•Aug 30, 2025•2 citations

Quantum machine learning with hybrid quantum-classical computation

Albert Cardenete Massip

This thesis explores variational quantum algorithms in order to solve optimization combinatorial problems using a meta-learning approach. This variational algorithm sare promising due to its capability to be used in the near future in the so-called NoisyIntermediate-Scale Quantum (NISQ) era, in which algorithms with high tolerance to noise could perform better than its classical counterparts. In this approach, a Recurrent Neural Network (RNN) optimizer tries to obtain the best set of parameters ofa quantum circuit for a given problem using the Qauntum Approximation Optimiza-tion Algorithm (QAOA), with the minimal amount of queries possible to a quantumcomputer. Thanks to the advances of parameter shift methods to compute the gradi-ents of parametrized quantum circuits, these algorithms could be even trained in realquantum computers without the need of highly demanding simulations. The resultspresented show that this approach is able to generalize to other problem instanceswhich the model has not seen before during training.

•Oct 1, 2020•0 citations

Improved Quantum Approximate Optimization Algorithm for Low‐Density Parity‐Check Channel Decoding

Han Zeng, Fan‐Xu Meng, Tian Luan +2 more

Abstract Quantum computing shows promise for 6G networks due to its parallel computing capabilities. In the context of the Noisy Intermediate‐Scale Quantum era, the introduction of hybrid quantum‐classical algorithms like Quantum Approximate Optimization Algorithm (QAOA) offer powerful solutions to many combinatorial optimization problems in 6G. This paper focuses on Low‐Density Parity‐Check (LDPC) channel decoding and proposes an improved QAOA algorithm assisted by the learning‐to‐learn strategy. We also investigate the parameter concentration phenomenon in QAOA‐based LDPC decoding to assess the rationality. To evaluate effectiveness, a comprehensive numerical expression for the energy expectation of single‐layer QAOA and propose indicators for transfer performance evaluation is provided. Based on simulation results, the similarity in parameter distribution across specific LDPC configurations is investigated. This similarity facilitates the transfer of training outcomes from smaller to larger‐scale problems for optimization initialization, thereby avoiding the need for retraining. This approach offers insights and potential solutions for rapid, large‐scale channel decoding in 6G networks, despite the current limitations of quantum hardware.

Advanced Quantum Technologies•Apr 14, 2024•5 citations

Reservoir Computing via Quantum Recurrent Neural Networks

Samuel Yen-Chi Chen, Daniel Fry, Deshmukh, Amol +2 more

Recent developments in quantum computing and machine learning have propelled the interdisciplinary study of quantum machine learning. Sequential modeling is an important task with high scientific and commercial value. Existing VQC or QNN-based methods require significant computational resources to perform the gradient-based optimization of a larger number of quantum circuit parameters. The major drawback is that such quantum gradient calculation requires a large amount of circuit evaluation, posing challenges in current near-term quantum hardware and simulation software. In this work, we approach sequential modeling by applying a reservoir computing (RC) framework to quantum recurrent neural networks (QRNN-RC) that are based on classical RNN, LSTM and GRU. The main idea to this RC approach is that the QRNN with randomly initialized weights is treated as a dynamical system and only the final classical linear layer is trained. Our numerical simulations show that the QRNN-RC can reach results comparable to fully trained QRNN models for several function approximation and time series prediction tasks. Since the QRNN training complexity is significantly reduced, the proposed model trains notably faster. In this work we also compare to corresponding classical RNN-based RC implementations and show that the quantum version learns faster by requiring fewer training epochs in most cases. Our results demonstrate a new possibility to utilize quantum neural network for sequential modeling with greater quantum hardware efficiency, an important design consideration for noisy intermediate-scale quantum (NISQ) computers.

arXiv:2211.02612•Nov 4, 2022•9 citations

Meta-Learning for Quantum Optimization via Quantum Sequence Model

Yu-Cheng Lin, Yu-Chao Hsu, Samuel Yen-Chi Chen

The Quantum Approximate Optimization Algorithm (QAOA) is a leading approach for solving combinatorial optimization problems on near-term quantum processors. However, finding good variational parameters remains a significant challenge due to the non-convex energy landscape, often resulting in slow convergence and poor solution quality. In this work, we propose a quantum meta-learning framework that trains advanced quantum sequence models to generate effective parameter initialization policies. We investigate four classical or quantum sequence models, including the Quantum Kernel-based Long Short-Term Memory (QK-LSTM), as learned optimizers in a "learning to learn" paradigm. Our numerical experiments on the Max-Cut problem demonstrate that the QK-LSTM optimizer achieves superior performance, obtaining the highest approximation ratios and exhibiting the fastest convergence rate across all tested problem sizes (n=10 to 13). Crucially, the QK-LSTM model achieves perfect parameter transferability by synthesizing a single, fixed set of near-optimal parameters, leading to a remarkable sustained acceleration of convergence even when generalizing to larger problems. This capability, enabled by the compact and expressive power of the quantum kernel architecture, underscores its effectiveness. The QK-LSTM, with only 43 trainable parameters, substantially outperforms the classical LSTM (56 parameters) and other quantum sequence models, establishing a robust pathway toward highly efficient parameter initialization for variational quantum algorithms in the NISQ era.

quant-pharXiv:2512.05058•International Conference on Quantum Communications, Networking, and Computing (QCNC)•Dec 5, 2025•0 citations

Meta-learning digitized-counterdiabatic quantum optimization

Pranav Chandarana, Pablo Suárez Vieites, Narendra N. Hegade +3 more

Abstract The use of variational quantum algorithms for optimization tasks has emerged as a crucial application for the current noisy intermediate-scale quantum computers. However, these algorithms face significant difficulties in finding suitable ansatz and appropriate initial parameters. In this paper, we employ meta-learning using recurrent neural networks to address these issues for the recently proposed digitized-counterdiabatic quantum approximate optimization algorithm (QAOA). By combining meta-learning and counterdiabaticity, we find suitable variational parameters and reduce the number of optimization iterations required. We demonstrate the effectiveness of our approach by applying it to the MaxCut problem and the Sherrington–Kirkpatrick model. Our method offers a short-depth circuit ansatz with optimal initial parameters, thus improving the performance of the state-of-the-art QAOA.

arXiv:2206.09966•Quantum Science and Technology•Jul 7, 2023•28 citations
With the development of 6G networks, the required computational power has been greatly increased, and quantum computing offers promising approaches for this purpose due to its properties such as superposition and entanglement. It has been proposed to combine the Quantum Approximation optimization Algorithm (QAOA) with the Learning-to-Learn (L2L) strategy using classical Recurrent Neural Networks (RNN) to solve the problem of channel decoding on Noisy IntermediateScale Quantum (NISQ) devices. In this paper, we build on this foundation to further investigate the impact of different RNNs on channel decoding. The experimental results show that by training the RNN model of Minimal Gated Unit (MGU) with fewer parameters, the initial parameters of the obtained QAOA have a certain degree of similarity with the Long Short Term Memory Network (LSTM), and the decoding results are comparable. The method of training RNNs improves the performance and saves the cost compared to the random initialization method. Despite the limitation by the current quantum hardware, this quantum computation approach provides the future for channel decoding in the 6G network a promising scheme.
Quantum Approximate Optimization Algorithms (QAOA) promise efficient solutions to classically intractable combinatorial optimization problems by harnessing shallowdepth quantum circuits. Yet, their performance and scalability often hinge on effective parameter optimization, which remains nontrivial due to rugged energy landscapes and hardware noise. In this work, we introduce a quantum meta-learning framework that combines quantum neural networks, specifically Quantum Long Short-Term Memory (QLSTM) architectures, with QAOA. By training the QLSTM optimizer on smaller graph instances, our approach rapidly generalizes to larger, more complex problems, substantially reducing the number of iterations required for convergence. Through comprehensive benchmarks on Max-Cut and Sherrington-Kirkpatrick model instances, we demonstrate that QLSTM-based optimizers converge faster and achieve higher approximation ratios compared to classical baselines, thereby offering a robust pathway toward scalable quantum optimization in the NISQ era.
This thesis explores variational quantum algorithms in order to solve optimization combinatorial problems using a meta-learning approach. This variational algorithm sare promising due to its capability to be used in the near future in the so-called NoisyIntermediate-Scale Quantum (NISQ) era, in which algorithms with high tolerance to noise could perform better than its classical counterparts. In this approach, a Recurrent Neural Network (RNN) optimizer tries to obtain the best set of parameters ofa quantum circuit for a given problem using the Qauntum Approximation Optimiza-tion Algorithm (QAOA), with the minimal amount of queries possible to a quantumcomputer. Thanks to the advances of parameter shift methods to compute the gradi-ents of parametrized quantum circuits, these algorithms could be even trained in realquantum computers without the need of highly demanding simulations. The resultspresented show that this approach is able to generalize to other problem instanceswhich the model has not seen before during training.
Abstract Quantum computing shows promise for 6G networks due to its parallel computing capabilities. In the context of the Noisy Intermediate‐Scale Quantum era, the introduction of hybrid quantum‐classical algorithms like Quantum Approximate Optimization Algorithm (QAOA) offer powerful solutions to many combinatorial optimization problems in 6G. This paper focuses on Low‐Density Parity‐Check (LDPC) channel decoding and proposes an improved QAOA algorithm assisted by the learning‐to‐learn strategy. We also investigate the parameter concentration phenomenon in QAOA‐based LDPC decoding to assess the rationality. To evaluate effectiveness, a comprehensive numerical expression for the energy expectation of single‐layer QAOA and propose indicators for transfer performance evaluation is provided. Based on simulation results, the similarity in parameter distribution across specific LDPC configurations is investigated. This similarity facilitates the transfer of training outcomes from smaller to larger‐scale problems for optimization initialization, thereby avoiding the need for retraining. This approach offers insights and potential solutions for rapid, large‐scale channel decoding in 6G networks, despite the current limitations of quantum hardware.
Recent developments in quantum computing and machine learning have propelled the interdisciplinary study of quantum machine learning. Sequential modeling is an important task with high scientific and commercial value. Existing VQC or QNN-based methods require significant computational resources to perform the gradient-based optimization of a larger number of quantum circuit parameters. The major drawback is that such quantum gradient calculation requires a large amount of circuit evaluation, posing challenges in current near-term quantum hardware and simulation software. In this work, we approach sequential modeling by applying a reservoir computing (RC) framework to quantum recurrent neural networks (QRNN-RC) that are based on classical RNN, LSTM and GRU. The main idea to this RC approach is that the QRNN with randomly initialized weights is treated as a dynamical system and only the final classical linear layer is trained. Our numerical simulations show that the QRNN-RC can reach results comparable to fully trained QRNN models for several function approximation and time series prediction tasks. Since the QRNN training complexity is significantly reduced, the proposed model trains notably faster. In this work we also compare to corresponding classical RNN-based RC implementations and show that the quantum version learns faster by requiring fewer training epochs in most cases. Our results demonstrate a new possibility to utilize quantum neural network for sequential modeling with greater quantum hardware efficiency, an important design consideration for noisy intermediate-scale quantum (NISQ) computers.
The Quantum Approximate Optimization Algorithm (QAOA) is a leading approach for solving combinatorial optimization problems on near-term quantum processors. However, finding good variational parameters remains a significant challenge due to the non-convex energy landscape, often resulting in slow convergence and poor solution quality. In this work, we propose a quantum meta-learning framework that trains advanced quantum sequence models to generate effective parameter initialization policies. We investigate four classical or quantum sequence models, including the Quantum Kernel-based Long Short-Term Memory (QK-LSTM), as learned optimizers in a "learning to learn" paradigm. Our numerical experiments on the Max-Cut problem demonstrate that the QK-LSTM optimizer achieves superior performance, obtaining the highest approximation ratios and exhibiting the fastest convergence rate across all tested problem sizes (n=10 to 13). Crucially, the QK-LSTM model achieves perfect parameter transferability by synthesizing a single, fixed set of near-optimal parameters, leading to a remarkable sustained acceleration of convergence even when generalizing to larger problems. This capability, enabled by the compact and expressive power of the quantum kernel architecture, underscores its effectiveness. The QK-LSTM, with only 43 trainable parameters, substantially outperforms the classical LSTM (56 parameters) and other quantum sequence models, establishing a robust pathway toward highly efficient parameter initialization for variational quantum algorithms in the NISQ era.
Abstract The use of variational quantum algorithms for optimization tasks has emerged as a crucial application for the current noisy intermediate-scale quantum computers. However, these algorithms face significant difficulties in finding suitable ansatz and appropriate initial parameters. In this paper, we employ meta-learning using recurrent neural networks to address these issues for the recently proposed digitized-counterdiabatic quantum approximate optimization algorithm (QAOA). By combining meta-learning and counterdiabaticity, we find suitable variational parameters and reduce the number of optimization iterations required. We demonstrate the effectiveness of our approach by applying it to the MaxCut problem and the Sherrington–Kirkpatrick model. Our method offers a short-depth circuit ansatz with optimal initial parameters, thus improving the performance of the state-of-the-art QAOA.