Searcharxiv⌕ Search

arXiv subjects

Muhammad Usman

Publications and source records attributed to Muhammad Usman.

At least 91 records · Page 5Linked to original sources

Fully convolutional 3D neural network decoders for surface codes with syndrome circuit noise

Artificial Neural Networks (ANNs) are a promising approach to the decoding problem of Quantum Error Correction (QEC), but have observed consistent difficulty when generalising performance to larger QEC codes. Recent scalability-focused approaches have split the decoding workload by using local ANNs to perform initial syndrome processing and leaving final processing to a global residual decoder. We investigated ANN surface code decoding under a scheme exploiting the spatiotemporal structure of syndrome data. In particular, we present a vectorised method for surface code data simulation and benchmark decoding performance when such data defines a multi-label classification problem and generative modelling problem for rotated surface codes with circuit noise after each gate and idle timestep. Performance was found to generalise to rotated surface codes of sizes up to $d=97$, with depolarisation parameter thresholds of up to $0.7\%$ achieved, competitive with h Minimum Weight Perfect Matching (MWPM). Improved latencies, compared with MWPM alone, were found starting at code distances of $d=33$ and $d=89$ under noise models above and below threshold respectively. These results suggest promising prospects for ANN-based frameworks for surface code decoding with performance sufficient to support the demands expected from fault-tolerant resource estimates.

quant-ph↗

Quantum Machine Learning

The meteoric rise of artificial intelligence in recent years has seen machine learning methods become ubiquitous in modern science, technology, and industry. Concurrently, the emergence of programmable quantum computers, coupled with the expectation that large-scale fault-tolerant machines will follow in the near to medium-term future, has led to much speculation about the prospect of quantum machine learning (QML), namely machine learning (ML) solutions which take advantage of quantum properties to outperform their classical counterparts. Indeed, QML is widely considered as one of the front-running use cases for quantum computing. In recent years, research in QML has gained significant global momentum. In this chapter, we introduce the fundamentals of QML and provide a brief overview of the recent progress and future trends in the field of QML. We highlight key opportunities for achieving quantum advantage in ML tasks, as well as describe some open challenges currently facing the field of QML. Specifically in the context of cybersecurity, we introduce the potential for QML in defence and security-sensitive applications, where it has been predicted that the seamless integration of quantum computing into ML will herald the development of robust and reliable QML systems, resilient against sophisticated threats arising from data manipulation and poisoning.

quant-ph↗

Context-Aware Unit Testing for Quantum Subroutines

Software testing is a critical component of the classical software development lifecycle, and this principle is expected to hold true for quantum software as it evolves toward large-scale production and adherence to industry standards. Developing and testing quantum software presents unique challenges due to the non-deterministic nature of quantum information, the high dimensionality of the underlying Hilbert space, complex hardware noise, and the inherent non-local properties of quantum systems. In this work, we model quantum subroutines as parametrized quantum channels and explore the feasibility of creating practical unit tests using probabilistic assertions, combined with either quantum tomography or statistical tests. To address the computational complexity associated with unit testing in quantum systems, we propose incorporating context-awareness into the testing process. The trade-offs between accuracy, state space coverage, and efficiency associated with the proposed theoretical framework for quantum unit testing have been demonstrated through its application to a simple three-qubit quantum subroutine that prepares a Greenberger-Horne-Zeilinger state, as well as to subroutines within a program implementing Shor's algorithm.

quant-ph↗

Multilingual Hate Speech Detection in Social Media Using Translation-Based Approaches with Large Language Models

Social media platforms are critical spaces for public discourse, shaping opinions and community dynamics, yet their widespread use has amplified harmful content, particularly hate speech, threatening online safety and inclusivity. While hate speech detection has been extensively studied in languages like English and Spanish, Urdu remains underexplored, especially using translation-based approaches. To address this gap, we introduce a trilingual dataset of 10,193 tweets in English (3,834 samples), Urdu (3,197 samples), and Spanish (3,162 samples), collected via keyword filtering, with a balanced distribution of 4,849 Hateful and 5,344 Not-Hateful labels. Our methodology leverages attention layers as a precursor to transformer-based models and large language models (LLMs), enhancing feature extraction for multilingual hate speech detection. For non-transformer models, we use TF-IDF for feature extraction. The dataset is benchmarked using state-of-the-art models, including GPT-3.5 Turbo and Qwen 2.5 72B, alongside traditional machine learning models like SVM and other transformers (e.g., BERT, RoBERTa). Three annotators, following rigorous guidelines, ensured high dataset quality, achieving a Fleiss' Kappa of 0.821. Our approach, integrating attention layers with GPT-3.5 Turbo and Qwen 2.5 72B, achieves strong performance, with macro F1 scores of 0.87 for English (GPT-3.5 Turbo), 0.85 for Spanish (GPT-3.5 Turbo), 0.81 for Urdu (Qwen 2.5 72B), and 0.88 for the joint multilingual model (Qwen 2.5 72B). These results reflect improvements of 8.75% in English (over SVM baseline 0.80), 8.97% in Spanish (over SVM baseline 0.78), 5.19% in Urdu (over SVM baseline 0.77), and 7.32% in the joint multilingual model (over SVM baseline 0.82). Our framework offers a robust solution for multilingual hate speech detection, fostering safer digital communities worldwide.

cs.CL↗

Quantum Kernel Learning for Small Dataset Modeling in Semiconductor Fabrication: Application to Ohmic Contact

Modeling complex semiconductor fabrication processes such as Ohmic contact formation remains challenging due to high-dimensional parameter spaces and limited experimental data. While classical machine learning (CML) approaches have been successful in many domains, their performance degrades in small-sample, nonlinear scenarios. In this work, we investigate quantum machine learning (QML) as an alternative, exploiting quantum kernels to capture intricate correlations from compact datasets. Using only 159 experimental GaN HEMT samples, we develop a quantum kernel-aligned regressor (QKAR) combining a shallow Pauli-Z feature map with a trainable quantum kernel alignment (QKA) layer. All models, including seven baseline CML regressors, are evaluated under a unified PCA-based preprocessing pipeline to ensure a fair comparison. QKAR consistently outperforms classical baselines across multiple metrics (MAE, MSE, RMSE), achieving a mean absolute error of 0.338 Omega mm when validated on experimental data. We further assess noise robustness and generalization through cross-validation and new device fabrication. These findings suggest that carefully constructed QML models could provide predictive advantages in data-constrained semiconductor modeling, offering a foundation for practical deployment on near-term quantum hardware. While challenges remain for both QML and CML, this study demonstrates QML's potential as a complementary approach in complex process modeling tasks.

cs.LG↗

InspectionV3: Enhancing Tobacco Quality Assessment with Deep Convolutional Neural Networks for Automated Workshop Management

The problems that tobacco workshops encounter include poor curing, inconsistencies in supplies, irregular scheduling, and a lack of oversight, all of which drive up expenses and worse quality. Large quantities make manual examination costly, sluggish, and unreliable. Deep convolutional neural networks have recently made strides in capabilities that transcend those of conventional methods. To effectively enhance them, nevertheless, extensive customization is needed to account for subtle variations in tobacco grade. This study introduces InspectionV3, an integrated solution for automated flue-cured tobacco grading that makes use of a customized deep convolutional neural network architecture. A scope that covers color, maturity, and curing subtleties is established via a labelled dataset consisting of 21,113 images spanning 20 quality classes. Expert annotators performed preprocessing on the tobacco leaf images, including cleaning, labelling, and augmentation. Multi-layer CNN factors use batch normalization to describe domain properties like as permeability and moisture spots, and so account for the subtleties of the workshop. Its expertise lies in converting visual patterns into useful information for enhancing workflow. Fast notifications are made possible by real-time, on-the-spot grading that matches human expertise. Images-powered analytics dashboards facilitate the tracking of yield projections, inventories, bottlenecks, and the optimization of data-driven choices. More labelled images are assimilated after further retraining, improving representational capacities and enabling adaptations for seasonal variability. Metrics demonstrate 97% accuracy, 95% precision and recall, 96% F1-score and AUC, 95% specificity; validating real-world viability.

cs.CV↗

USEFUSE: Uniform Stride for Enhanced Performance in Fused Layer Architecture of Deep Neural Networks

Convolutional Neural Networks (CNNs) are crucial in various applications, but their deployment on resource-constrained edge devices poses challenges. This study presents the Sum-of-Products (SOP) units for convolution, which utilize low-latency left-to-right bit-serial arithmetic to minimize response time and enhance overall performance. The study proposes a methodology for fusing multiple convolution layers to reduce off-chip memory communication and increase overall performance. An effective mechanism detects and skips inefficient convolutions after ReLU layers, minimizing power consumption without compromising accuracy. Furthermore, efficient tile movement guarantees uniform access to the fusion pyramid. An analysis demonstrates the utile stride strategy improves operational intensity. Two designs cater to varied demands: one focuses on minimal response time for mission-critical applications, and another focuses on resource-constrained devices with comparable latency. This approach notably reduced redundant computations, improving the efficiency of CNN deployment on edge devices.

cs.LG↗

Quantum Generative Learning for High-Resolution Medical Image Generation

Integration of quantum computing in generative machine learning models has the potential to offer benefits such as training speed-up and superior feature extraction. However, the existing quantum generative adversarial networks (QGANs) fail to generate high-quality images due to their patch-based, pixel-wise learning approaches. These methods capture only local details, ignoring the global structure and semantic information of images. In this work, we address these challenges by proposing a quantum image generative learning (QIGL) approach for high-quality medical image generation. Our proposed quantum generator leverages variational quantum circuit approach addressing scalability issues by extracting principal components from the images instead of dividing them into patches. Additionally, we integrate the Wasserstein distance within the QIGL framework to generate a diverse set of medical samples. Through a systematic set of simulations on X-ray images from knee osteoarthritis and medical MNIST datasets, our model demonstrates superior performance, achieving the lowest Fréchet Inception Distance (FID) scores compared to its classical counterpart and advanced QGAN models reported in the literature.

quant-ph↗

Stabilizer Tensor Networks with Magic State Injection

This work augments the recently introduced Stabilizer Tensor Network (STN) protocol with magic state injection, reporting a new framework with significantly enhanced ability to simulate circuits with an extensive number of non-Clifford operations. Specifically, for random $T$-doped $N$-qubit Clifford circuits the computational cost of circuits prepared with magic state injection scales as $\mathcal{O}(\text{poly}(N))$ when the circuit has $t \lesssim N$ $T$-gates compared to an exponential scaling for the STN approach, which is demonstrated in systems of up to $200$ qubits. In the case of the Hidden Bit Shift circuit, a paradigmatic benchmarking system for extended stabilizer methods with a tunable amount of magic, we report that our magic state injected STN framework can efficiently simulate $4000$ qubits and $320$ $T$-gates. These findings provide a promising outlook for the use of this protocol in the classical modelling of quantum circuits that are conventionally difficult to simulate efficiently.

quant-ph↗

Classical Autoencoder Distillation of Quantum Adversarial Manipulations

Quantum neural networks have been proven robust against classical adversarial attacks, but their vulnerability against quantum adversarial attacks is still a challenging problem. Here we report a new technique for the distillation of quantum manipulated image datasets by using classical autoencoders. Our technique recovers quantum classifier accuracies when tested under standard machine learning benchmarks utilising MNIST and FMNIST image datasets, and PGD and FGSM adversarial attack settings. Our work highlights a promising pathway to achieve fully robust quantum machine learning in both classical and quantum adversarial scenarios.

quant-ph↗

Variational Quantum Machine Learning with Quantum Error Detection

Quantum machine learning (QML) is an emerging field that promises advantages such as faster training, improved reliability and superior feature extraction over classical counterparts. However, its implementation on quantum hardware is challenging due to the noise inherent in these systems, necessitating the use of quantum error correction (QEC) codes. Current QML research remains primarily theoretical, often assuming noise-free environments and offering little insight into the integration of QEC with QML implementations. To address this, we investigate the performance of a simple, parity-classifying Variational Quantum Classifier (VQC) implemented with the [[4,2,2]] error-detecting stabiliser code in a simulated noisy environment, marking the first study into the implementation of a QML algorithm with a QEC code. We invoke ancilla qubits to logically encode rotation gates, and classically simulate the logically-encoded VQC under two simple noise models representing gate noise and environmental noise. We demonstrate that the stabiliser code improves the training accuracy at convergence compared to noisy implementations without QEC. However, we find that the effectiveness and reliability of error detection is contingent upon keeping the ancilla qubit error rates below a specific threshold, due to the propagation of ancilla errors to the physical qubits. Our results provide an important insight: for QML implementations with QEC codes that both require ancilla qubits for logical rotations and cannot fully correct errors propagated between ancilla and physical qubits, the maximum achievable accuracy of the QML model is limited. This highlights the need for additional error correction or mitigation strategies to support the practical implementation of QML algorithms with QEC on quantum devices.

quant-ph↗

Fast and Noise-aware Machine Learning Variational Quantum Eigensolver Optimiser

The Variational Quantum Eigensolver (VQE) is a hybrid quantum-classical algorithm for preparing ground states in the current era of noisy devices. The classical component of the algorithm requires a large number of measurements on intermediate parameter values that are typically discarded. However, intermediate steps across many calculations can contain valuable information about the relationship between the quantum circuit parameters, resultant measurements, and noise specific to the device. In this work, we use supervised machine learning on the intermediate parameter and measurement data to predict optimal final parameters. Our technique optimises parameters leading to chemically accurate ground state energies much faster than conventional techniques. It requires significantly fewer iterations and simultaneously shows resilience to coherent errors if trained on noisy devices. We demonstrate this technique on IBM quantum devices by predicting ground state energies of H$_2$ for one and two qubits; H$_3$ for three qubits; and HeH$^+$ for four qubits where it finds optimal angles using only modeled data for training.

quant-ph↗

High Performance 5G FR-2 Millimeter-Wave Antenna Array for Point-to-Point and Point-to-Multipoint Operation: Design and OTA Measurements Using a Compact Antenna Test Range

This paper presents the design and comprehensive measurements of a high-performance 8-element linear array and a compact high-gain 32-element planar antenna array covering the n257 (26.5--29.5 GHz) FR-2 millimeter-wave (mmWave) band. First, an 8-element series-fed linear array is designed with a fan-shaped pattern for 5G point-to-multipoint connectivity. Then a 4-way corporate-series feed network is designed for a high-gain 32-element compact and directive array for point-to-point mmWave connectivity. Comprehensive over-the-air (OTA) measurements are conducted using a state-of-the-art compact antenna test range (CATR) system, enabling precise characterization of radiation patterns across a 180^\circ span in the azimuth and elevation planes. The planar array achieves a peak measured gain of 18.45 dBi at 28.5 GHz, with half-power beamwidths ranging from 11^\circ--13^\circ (wide axis) and 23^\circ--27^\circ (narrow axis) across the band of interest. The sidelobe levels are below -10 dB in the desired band of interest. The measured results match well with the simulation results. The designed antenna array is applicable to various emerging 5G and beyond mmWave applications such as high data rate mmWave wireless backhaul, mmWave near-field focusing, high-resolution indoor radar systems, 28 GHz Local Multipoint Distribution Service (LMDS), as well as the characterization of mmWave path loss and channel sounding in diverse indoor environments.

eess.SP↗

Nanowire design by deep learning for energy efficient photonic technologies

This work describes our vision and proposal for the design of next generation photonic devices based on custom-designed semiconductor nanowires. The integration of multi-million-atom electronic structure and optical simulations with the supervised machine learning models will pave the way for transformative nanowire-based technologies, offering opportunities for the next generation energy-efficient greener photonics.

physics.optics↗

Automatizing the search for mass resonances using BumpNet

The search for resonant mass bumps in invariant-mass distributions remains a cornerstone strategy for uncovering Beyond the Standard Model (BSM) physics at the Large Hadron Collider (LHC). Traditional methods often rely on predefined functional forms and exhaustive computational and human resources, limiting the scope of tested final states and selections. This work presents BumpNet, a machine learning-based approach leveraging advanced neural network architectures to generalize and enhance the Data-Directed Paradigm (DDP) for resonance searches. Trained on a diverse dataset of smoothly-falling analytical functions and realistic simulated data, BumpNet efficiently predicts statistical significance distributions across varying histogram configurations, including those derived from LHC-like conditions. The network's performance is validated against idealized likelihood ratio-based tests, showing minimal bias and strong sensitivity in detecting mass bumps across a range of scenarios. Additionally, BumpNet's application to realistic BSM scenarios highlights its capability to identify subtle signals while managing the look-elsewhere effect. These results underscore BumpNet's potential to expand the reach of resonance searches, paving the way for more comprehensive explorations of LHC data in future analyses.

physics.data-an↗

DSLR-CNN: Efficient CNN Acceleration using Digit-Serial Left-to-Right Arithmetic

Digit-serial arithmetic has emerged as a viable approach for designing hardware accelerators, reducing interconnections, area utilization, and power consumption. However, conventional methods suffer from performance and latency issues. To address these challenges, we propose an accelerator design using left-to-right (LR) arithmetic, which performs computations in a most-significant digit first (MSDF) manner, enabling digit-level pipelining. This leads to substantial performance improvements and reduced latency. The processing engine is designed for convolutional neural networks (CNNs), which includes low-latency LR multipliers and adders for digit-level parallelism. The proposed DSLR-CNN is implemented in Verilog and synthesized with Synopsys design compiler using GSCL 45nm technology, the DSLR-CNN accelerator was evaluated on AlexNet, VGG-16, and ResNet-18 networks. Results show significant improvements across key performance evaluation metrics, including response time, peak performance, power consumption, operational intensity, area efficiency, and energy efficiency. The peak performance measured in GOPS of the proposed design is 4.37x to 569.11x higher than contemporary designs, and it achieved 3.58x to 44.75x higher peak energy efficiency (TOPS/W), outperforming conventional bit-serial designs.

cs.AR↗

Magic State Injection on IBM Quantum Processors Above the Distillation Threshold

The surface code family is a promising approach to implementing fault-tolerant quantum computations. Universal fault-tolerance requires error-corrected non-Clifford operations, in addition to Clifford gates, and for the former, it is imperative to experimentally demonstrate additional resources known as magic states. Another challenge is to efficiently embed surface codes into quantum hardware with connectivity constraints. This work simultaneously addresses both challenges by employing a qubit-efficient rotated heavy-hexagonal surface code for IBM quantum processors (\texttt{ibm\_fez}) and implementing the magic state injection protocol. Our work reports error thresholds for both logical bit- and phase-flip errors, of $\approx0.37\%$ and $\approx0.31\%$, respectively, which are higher than the threshold values previously reported with traditional embedding. The post-selection-based preparation of logical magic states $|H_L\rangle$ and $|T_L\rangle$ achieve fidelities of $0.8806\pm0.0002$ and $0.8665\pm0.0003$, respectively, which are both above the magic state distillation threshold. Additionally, we report the minimum fidelity among injected arbitrary single logical qubit states as $0.8356\pm0.0003$. Our work demonstrates the potential for realising non-Clifford logical gates by producing high-fidelity logical magic states on IBM quantum devices.

quant-ph↗

Architectural Patterns for Designing Quantum Artificial Intelligence Systems

Utilising quantum computing technology to enhance artificial intelligence systems is expected to improve training and inference times, increase robustness against noise and adversarial attacks, and reduce the number of parameters without compromising accuracy. However, moving beyond proof-of-concept or simulations to develop practical applications of these systems while ensuring high software quality faces significant challenges due to the limitations of quantum hardware and the underdeveloped knowledge base in software engineering for such systems. In this work, we have conducted a systematic mapping study to identify the challenges and solutions associated with the software architecture of quantum-enhanced artificial intelligence systems. The results of the systematic mapping study reveal several architectural patterns that describe how quantum components can be integrated into inference engines, as well as middleware patterns that facilitate communication between classical and quantum components. Each pattern realises a trade-off between various software quality attributes, such as efficiency, scalability, trainability, simplicity, portability, and deployability. The outcomes of this work have been compiled into a catalogue of architectural patterns.

cs.SE↗