Quantum Algorithm for Distributed Reduction of Entanglements (QADR): A Trainable and Simulation-Efficient QML Framework
For quantum machine learning practitioners, QADR offers a practical solution to scalability and trainability issues in NISQ-era VQCs, enabling larger problem sizes.
QADR decomposes global VQCs into localized sub-circuits, reducing classical simulation memory from O(2^n) to O(n·2^{2d+1}) and mitigating barren plateaus. It scales to 2000 features where standard VQCs fail, matching classical performance on MNIST and NASA IMS tasks.
Training Variational Quantum Circuits (VQCs) under Noisy Intermediate-Scale Quantum (NISQ) constraints introduces severe computational limitations: classical statevector simulation memory scales exponentially ($\mathcal{O}(2^n)$), and global cost functions suffer from barren plateaus where gradient variance decays exponentially ($\mathcal{O}(1/2^n)$). This paper introduces and evaluates the Quantum Algorithm for Distributed Reduction of Entanglements (QADR), a hybrid quantum-classical machine learning framework that decomposes a global $n$-qubit VQC into localized sub-circuits operating approximately within the causal light cones of individual target qubits. QADR reduces classical simulation memory scaling from $\mathcal{O}(2^n)$ to $\mathcal{O}(n \cdot 2^{2d+1})$ for a light cone radius $d$, while naturally mitigating global barren plateaus. We benchmark QADR against standard global VQCs, Support Vector Machines (SVM), and two customized classical parameter-matched neural networks (CANN and PMNN) on the MNIST dataset and the high-dimensional NASA IMS wind turbine drivetrain diagnostic task. QADR demonstrates excellent scalability, operating successfully at $n_{\text{features}}=2000$ where standard global VQCs crash due to memory exhaustion, while matching or exceeding the performance of optimized classical architectures.