HQCC: A Hybrid Quantum-Classical Classifier with Adaptive Structure

2025-04-02

AI Summary

To address the performance limitations of quantum machine learning (QML) on Noisy Intermediate-Scale Quantum (NISQ) devices caused by fixed-architecture parameterized quantum circuits (PQCs), this paper proposes HQCC, an adaptive-architecture hybrid quantum-classical classifier. Methodologically, it introduces—(i) an LSTM-driven dynamic circuit generation mechanism enabling real-time PQC topology evolution; (ii) a local quantum filter to enhance scalability and robustness in quantum feature extraction; and (iii) an architecture plasticity mechanism that dynamically balances entanglement depth against noise resilience. Evaluated on the TensorCircuit platform, HQCC achieves 97.12% accuracy on MNIST and significantly outperforms state-of-the-art baselines on Fashion-MNIST. This work constitutes the first integration of neural-network-driven, adaptive circuit architecture design into QML, establishing a novel paradigm for practical quantum learning on NISQ hardware.

Technology Category

• Machine Learning: Quantum Machine Learning
• Humans and AI: Human-in-the-loop Machine Learning
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• Security and Privacy: Security and privacy of machine learning and AI applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Parameterized Quantum Circuits (PQCs) with fixed structures severely degrade the performance of Quantum Machine Learning (QML). To address this, a Hybrid Quantum-Classical Classifier (HQCC) is proposed. It opens a practical way to advance QML in the Noisy Intermediate-Scale Quantum (NISQ) era by adaptively optimizing the PQC through a Long Short-Term Memory (LSTM) driven dynamic circuit generator, utilizing a local quantum filter for scalable feature extraction, and exploiting architectural plasticity to balance the entanglement depth and noise robustness. We realize the HQCC on the TensorCircuit platform and run simulations on the MNIST and Fashion MNIST datasets, achieving up to 97.12% accuracy on MNIST and outperforming several alternative methods.

Problem

• Adaptively optimizing PQC to enhance Quantum Machine Learning performance
• Using LSTM-driven dynamic circuit generator for noise robustness
• Achieving high accuracy on MNIST and Fashion MNIST datasets

Innovation

• Adaptive PQC optimization via LSTM-driven generator
• Local quantum filter for scalable feature extraction
• Architectural plasticity balances entanglement and noise