Quantum computers in the current noisy intermediate-scale quantum (NISQ) era face two major limitations - size and error vulnerability. Although quantum error correction (QEC) methods exist, they are not applicable at the current size of computers, requiring thousands of qubits, while NISQ systems have nearly one hundred at most. One common approach to improve reliability is to adjust the compilation process to create a more reliable final circuit, where the two most critical compilation decisions are the qubit allocation and qubit routing problems. We focus on solving the qubit allocation problem and identifying initial layouts that result in a reduction of error. To identify these layouts, we combine reinforcement learning with a graph neural network (GNN)-based Q-network to process the mesh topology of the quantum computer, known as the backend, and make mapping decisions, creating a Graph Neural Network Assisted Quantum Compilation (GNAQC) strategy. We train the architecture using a set of four backends and six circuits and find that GNAQC improves output fidelity by roughly 12.7% over pre-existing allocation methods.
Quantum computers in the current noisy intermediate-scale quantum (NISQ) era face two major limitations - size and error vulnerability. Although quantum error correction (QEC) methods exist, they are not applicable at the current size of computers, requiring thousands of qubits, while NISQ systems have nearly one hundred at most. One common approach to improve reliability is to adjust the compilation process to create a more reliable final circuit, where the two most critical compilation decisions are the qubit allocation and qubit routing problems. We focus on solving the qubit allocation problem and identifying initial layouts that result in a reduction of error. To identify these layouts, we combine reinforcement learning with a graph neural network (GNN)-based Q-network to process the mesh topology of the quantum computer, known as the backend, and make mapping decisions, creating a Graph Neural Network Assisted Quantum Compilation (GNAQC) strategy. We train the architecture using a set of four backends and six circuits and find that GNAQC improves output fidelity by roughly 12.7% over pre-existing allocation methods.
Quantum computing is a quickly growing field with great potential for future technology. Quantum computers in the current noisy intermediate-scale quantum (NISQ) era face two major limitations - qubit count and error vulnerability. Although quantum error correction (QEC) methods exist, they are not applicable to the current size of computers, requiring thousands of qubits, while current NISQ systems have hundreds at most. It is therefore imperative to improve the reliability of the circuits as much as possible to make them robust to the errors that will occur. One common approach is to adjust the compilation process of a circuit to create a final circuit with improved reliability. However, there are many decisions to be made when compiling that affect the final performance of the circuit, two of the most critical ones being the mapping of logical to physical qubits (the qubit allocation problem) and the movement of qubits to satisfy two-qubit gate adjacency requirements (the qubit routing problem). We focus on solving the qubit allocation problem and identifying initial layouts that reduce error. To identify these layouts, we combine reinforcement learning with a graph neural network (GNN)-based Q-network for analyzing both the connections and error rates of the graph-like backend of superconducting quantum computers to make mapping decisions, creating a Graph Neural Network Assisted Compilation (GNAQC) strategy. We provide both the circuit and the properties of the target backend as input to guide the decision-making process. We work with the IBM Qiskit API to compile and simulate our quantum circuits. We train the architecture using a set of four backends and six circuits and find that GNAQC generally outperforms pre-existing qubit allocation algorithms, increasing final relative output fidelity by roughly 12.7%.
Modern computers can experience a variety of transient errors due to the surrounding environment, known as soft faults. Although the frequency of these faults is low enough to not be noticeable on personal computers, they become a considerable concern during large-scale distributed computations or systems in more vulnerable environments like satellites. These faults occur as a bit flip of some value in a register, operation, or memory during execution. They surface as either program crashes, hangs, or silent data corruption (SDC), each of which can waste time, money, and resources. Hardware methods, such as shielding or error correcting memory (ECM), exist, though they can be difficult to implement, expensive, and may be limited to only protecting against errors in specific locations. Researchers have been exploring software detection and correction methods as an alternative, commonly trading either overhead in execution time or memory usage to protect against faults. Quantum computers, a relatively recent advancement in computing technology, experience similar errors on a much more severe scale. The errors are more frequent, costly, and difficult to detect and correct. Error correction algorithms like Shor’s code promise to completely remove errors, but they cannot be implemented on current noisy intermediate-scale quantum (NISQ) systems due to the low number of available qubits. Until the physical systems become large enough to support error correction, researchers instead have been studying other methods to reduce and compensate for errors. In this work, we present two methods for improving the resilience of classical processes, both single- and multi-threaded. We then introduce quantum computing and compare the nature of errors and correction methods to previous classical methods. We further discuss two designs for improving compilation of quantum circuits. One method, focused on quantum neural networks (QNNs), takes advantage of partial compilation to avoid recompiling the entire circuit each time. The other method is a new approach to compiling quantum circuits using graph neural networks (GNNs) to improve the resilience of quantum circuits and increase fidelity. By using GNNs with reinforcement learning, we can train a compiler to provide improved qubit allocation that improves the success rate of quantum circuits.
To generate physical quantum circuits, quantum gates are often added to the quantum circuit to satisfy the hardware constraints. This process is called quantum circuit mapping. Noise-aware mapping techniques generate physical quantum circuits for noisy intermediate-scale quantum (NISQ) computers. However, the absence of an accurate noise model of the quantum hardware limits its performance. In this article, we propose a noise adaptive quantum circuit mapping approach using reinforcement learning (RL) and graph neural network (GNN)-based reliability predictor. Our RL agent learns a quantum circuit mapping policy that significantly improves the quantum circuit output fidelity by interacting with the environment, which adopts a GNN reliability model that acts as quantum hardware and estimates the physical quantum circuit fidelity. Furthermore, we propose a multi-GNN reliability model to speed up the inference while maintaining high accuracy. Our proposed RL framework fills the gap between the simplified reliability models of the quantum hardware and the realistic noise impact of the quantum hardware on the quantum circuits. We demonstrate the improvement in the output fidelity of quantum circuits generated using our approach compared to other quantum circuit mapping techniques across different real-world quantum hardware using multiple-seed experiments.
Abstract Logical-qubit allocation is a central compilation problem for fault-tolerant quantum architectures, where circuit qubits must be assigned to feasible chip tiles while minimizing the ancilla-qubit cost (the number of extra workspace qubits consumed by the routing corridors required to execute the circuit) and preserving efficient access to magic-state resources. This hard combinatorial problem has resisted full automation: existing allocators struggle to jointly capture global circuit interactions, tile-level routing feasibility, and generalization across circuit families. We propose a graph-based reinforcement-learning framework in which a graph neural network is first pretrained to predict the ancilla-qubit cost of circuit-allocation pairs from allocation-aware circuit graphs (graphs that combine circuit structure with the current placement of each qubit), and the resulting encoder is transferred to a reinforcement learning agent that assigns tiles one at a time. As a result, the agent learns to utilize past experience to produce higher-quality solutions to new instances of the allocation problem. Evaluated on MQTBench circuits compiled to lattice-surgery instructions under leave-one-family-out cross-validation, our allocator reduces average ancilla-qubit cost by 36.7% compared to the state-of-the-art ECMAS+ baseline and achieves lower cost in 57 out of 64 qubit-size bins. The method further exhibits family-dependent scale extrapolation and responsive few-shot adaptation, establishing learned logical-qubit allocation as a scalable and transferable paradigm that outperforms ECMAS+ under the evaluated fixed-chip setting.
Abstract Modular quantum architectures have emerged as a promising approach for scaling quantum computing systems by connecting multiple Quantum Processing Units (QPUs). However, this approach introduces significant challenges due to costly inter-core operations between chips and quantum state transfers, which contribute to noise and quantum decoherence. This paper presents QARMA, a novel Qubit mapping using Attention-based deep Reinforcement learning (DRL) for Modular quantum Architectures, along with its extension QARMA-R that incorporates dynamic qubit reuse capabilities. Our approach combines an attention-based mechanism with Graph Neural Networks (GNN) to learn optimal qubit allocation, routing, and reuse strategies that minimize inter-core communications.Our experimental results show significant improvements over state-of-the-art approaches. Compared to highly-optimized Qiskit with modular architecture configuration, QARMA-R reduces inter-core communications by up to 100% (on average 86%), while QARMA maintains 15-40% improvement for larger circuits without reuse. Against traditional modular qubit mapping, our approach achieves 97-100% reduction in inter-core operation.The proposed methods advance quantum circuit compilation techniques and enable the execution of more extensive quantum algorithms on resource-constrained modular quantum systems, contributing to the growing body of research on scalable quantum computing architectures.
The current advancement in quantum computers has been focusing on increasing the number of qubits and enhancing their fidelity. However, the available quantum devices, known as intermediate scale quantum (NISQ) computers, still suffer from different sources of noise that impact their reliability. Thus, practical noise modeling is of great importance in the development of quantum error mitigation approaches. In this article, we propose a machine learning (ML)-based scheme to predict the output fidelity of the quantum circuit executed on NISQ devices. We show the benefit of using graph neural network (GNN)-based models compared to traditional ML-based models in capturing the quantum circuit structure in addition to its gates’ features, which enable characterizing unpredicted quantum circuit errors. We use different metrics to measure the fidelity of the quantum circuit output. Our experimental results using different quantum algorithms executed on IBM Q Guadalupe quantum computer show the high prediction accuracy of our ML reliability models. Our results also show that our models can guide the single-qubit gate rescheduling to improve the output fidelity of the quantum circuit without the need for prior execution of dedicated calibration circuits.