Quantum-Based Self-Attention Mechanism for Hardware-Aware Differentiable Quantum Architecture Search

TL;DR

The paper tackles automated design of parameterized quantum circuits under NISQ hardware noise by introducing QBSA-DQAS, a quantum-native self-attention mechanism integrated into differentiable quantum architecture search with hardware-aware multi-objective optimization. It replaces classical similarity with quantum-derived attention, optimizes for noisy expressibility and Probability of Successful Trials, and applies a post-search circuit simplification to yield compact, hardware-ready PQCs. Empirical validation across VQE tasks and large-scale Wireless Sensor Networks demonstrates higher accuracy in noiseless and noisy settings and substantial energy savings, showing the method's robustness and practical relevance for near-term quantum devices. The framework offers a scalable, transferable approach to quantum architecture search across domains and lays groundwork for broader generalization studies.

Abstract

The automated design of parameterized quantum circuits for variational algorithms in the NISQ era faces a fundamental limitation, as conventional differentiable architecture search relies on classical models that fail to adequately represent quantum gate interactions under hardware noise. We introduce the Quantum-Based Self-Attention for Differentiable Quantum Architecture Search (QBSA-DQAS), a meta-learning framework featuring quantum-based self-attention and hardware-aware multi-objective search. The framework employs a two-stage quantum self-attention module that computes contextual dependencies by mapping architectural parameters through parameterized quantum circuits, replacing classical similarity metrics with quantum-derived attention scores, then applies position-wise quantum transformations for feature enrichment. Architecture search is guided by a task-agnostic multi-objective function jointly optimizing noisy expressibility and Probability of Successful Trials (PST). A post-search optimization stage applies gate commutation, fusion, and elimination to reduce circuit complexity. Experimental validation demonstrates superior performance on VQE tasks and large-scale Wireless Sensor Networks. For VQE on H$_2$, QBSA-DQAS achieves 0.9 accuracy compared to 0.89 for standard DQAS. Post-search optimization reduces discovered circuit complexity by up to 44% in gate count and 47% in depth without accuracy degradation. The framework maintains robust performance across three molecules and five IBM quantum hardware noise models. For WSN routing, discovered circuits achieve 8.6% energy reduction versus QAOA and 40.7% versus classical greedy methods, establishing the effectiveness of quantum-native architecture search for NISQ applications.

Figures (9)

• Figure 1: An overview of the QBSA-DQAS architecture search algorithm.
• Figure 2: Quantum feature map circuit for attention.
• Figure 3: Position-wise feed-forward quantum circuit.
• Figure 4: The post-search optimization cascade.
• Figure 5: VQE accuracy comparison for the $H_2$ molecule in a noiseless environment.