Quantum Patches for Efficient Learning

Mahua Nandy Pal, Minakshi Banerjee, Ankit Sarkar

Abstract Deep neural networks were previously used in the arena of image retrieval. Siamese network architecture is also used for image similarity comparison. Recently, the application of quantum computing in different fields has gained research interest. Researchers are keen to explore the prospect of quantum circuit implementation in terms of supervised learning, resource utilization, and energy‐efficient reversible computing. In this study, the authors propose an application of quantum circuit in Siamese architecture and explored its efficiency in the field of exudate‐affected retinal image patch retrieval. Quantum computing applied within Siamese network architecture may be effective for image patch characteristic comparison and retrieval work. Although there is a restriction of managing high‐dimensional inner product space, the circuit with a limited number of qubits represents exudate‐affected retinal image patches and retrieves similar patches from the patch database. Parameterized quantum circuit (PQC) is implemented using a quantum machine learning library on Google Cirq framework. PQC model is composed of classical pre/post‐processing and parameterized quantum circuit. System efficiency is evaluated with the most widely used retrieval evaluation metrics: mean average precision (MAP) and mean reciprocal rank (MRR). The system achieved an encouraging and promising result of 98.1336% MAP and 100% MRR. Image pixels are implicitly converted to rectangular grid qubits in this experiment. The experimentation was further extended to IBM Qiskit framework also. In Qiskit, individual pixels are explicitly encoded using novel enhanced quantum representation (NEQR) image encoding algorithm. The probability distributions of both query and database patches are compared through Jeffreys distance to retrieve similar patches.

Ryan Wilkinson

Maria Schuld, Francesco Petruccione

Maria Francisca Madeira, Alessandro Poggiali, Jeanette Miriam Lorenz

Quantum Machine Learning investigates the pos-sibility of quantum computers enhancing Machine Learning algorithms. Anomaly segmentation is a fundamental task in various domains to identify irregularities at sample level and can be addressed with both supervised and unsupervised methods. Autoencoders are commonly used in unsupervised tasks, where models are trained to reconstruct normal instances efficiently, al-lowing anomaly identification through high reconstruction errors. While quantum autoencoders have been proposed in the literature, their application to anomaly segmentation tasks remains unexplored. In this paper, we introduce a patch-based quantum autoencoder (QPB-AE) for image anomaly segmentation, with a number of parameters scaling logarithmically with patch size. QPB-AE reconstructs the quantum state of the embedded input patches, computing an anomaly map directly from measurement through a SWAP test without reconstructing the input image. We evaluate its performance across multiple datasets and parameter configurations and compare it against a classical counterpart.

Al Amin, Rakib Hossain, Mahbubul Islam +2 more

Quantum machine learning has emerged as a promising tool for pattern recognition, yet many audio-focused approaches still treat spectrograms as generic images and do not explicitly exploit their time-frequency structure. We propose Q-Patch, a quantum feature map tailored to audio that encodes local time-frequency patches from mel-spectrograms into quantum states using shallow, hardware-efficient circuits with adjacency-aware entanglement. Each selected patch is summarized by a compact four-dimensional acoustic descriptor and mapped to a four-qubit circuit with depth at most three, enabling practical quantum kernel construction under near-term constraints. We evaluate Q-Patch on an audio spoofing detection task using a controlled, balanced protocol and compare it with size-matched classical baselines. Q-Patch improves discrimination between bona fide and spoofed samples, achieving an area under the receiver operating characteristic curve (AUROC) of 0.87, compared with 0.82 for a radial basis function support vector machine (RBF-SVM) trained on the same patch-level features. Kernel-space analysis further reveals a clear class structure, with cross-class similarity around 0.615 and within-class self-similarity of 1.00. Overall, Q-Patch provides a practical framework for incorporating time-frequency-aware representations into quantum kernel learning for audio authenticity assessment in low-resource settings.

Pawan Whig, Pavika Sharma, Ahmad A. Elngar +1 more