Searcharxiv⌕ Search

arXiv subjects

Muhammad Usman

Publications and source records attributed to Muhammad Usman.

At least 55 records · Page 3Linked to original sources

Prophecy: Inferring Formal Properties from Neuron Activations

We present Prophecy, a tool for automatically inferring formal properties of feed-forward neural networks. Prophecy is based on the observation that a significant part of the logic of feed-forward networks is captured in the activation status of the neurons at inner layers. Prophecy works by extracting rules based on neuron activations (values or on/off statuses) as preconditions that imply certain desirable output property, e.g., the prediction being a certain class. These rules represent network properties captured in the hidden layers that imply the desired output behavior. We present the architecture of the tool, highlight its features and demonstrate its usage on different types of models and output properties. We present an overview of its applications, such as inferring and proving formal explanations of neural networks, compositional verification, run-time monitoring, repair, and others. We also show novel results highlighting its potential in the era of large vision-language models.

cs.LG↗

Efficient Quantum Algorithms for Higher-Order Coupled Oscillators

Higher-order networks with multiway interactions can exhibit collective dynamical phenomena that are absent in traditional pairwise network models. However, analyzing such dynamics becomes computationally prohibitive as their state space grows combinatorially in the multiway interaction order. Here we develop quantum algorithms for two central tasks -- synchronization estimation and certification of the no-phase-locking regime -- in the simplicial Kuramoto model. This model is a higher-order generalization of the celebrated Kuramoto model for coupled oscillators on graph-based networks. Under explicit assumptions on data access and types, and simplicial structure, we derive end-to-end quantum gate complexities and identify regimes with polynomial quantum advantage for synchronization estimation and super-polynomial quantum advantage for no-phase-locking certification over classical methods. More broadly, these results extend quantum algorithms for higher-order networks from structural analysis to nonlinear dynamical diagnostics, easing a major computational bottleneck and opening a route to quantum methods for probing higher-order phenomena beyond the reach of direct classical approaches.

quant-ph↗

Feature-level analysis and adversarial transfer in rotationally equivariant quantum machine learning

Group-equivariant quantum models are designed to exploit symmetry and can improve trainability, but it remains unclear how symmetry constraints shape their adversarial robustness. We study this question through a feature-level analysis of equivariant quantum models in a transfer-attack setting. Under equivariance with an invariant readout, predictions depend only on the group-twirled input, which identifies the symmetry-invariant information accessible to the model together with a complementary uninformative subspace. Specializing this framework to a rotationally equivariant quantum model, we derive an explicit characterization of the accessible information in terms of rotation-invariant image statistics distributed across distinct symmetry sectors. Using targeted input transformations, we determine which of these statistics are actually relied upon for classification across several datasets. We find that equivariance alone does not guarantee transfer robustness: even within the restricted invariant feature space, the model can rely on brittle statistics, particularly ring-averaged intensities in the rotationally equivariant model, that remain vulnerable to classical transfer attacks. Guided by this analysis, we show that suppressing the symmetry sector associated with the brittle feature substantially improves robustness. These results establish a systematic mechanism to exploit symmetry-dependent features for adversarial robustness in future quantum machine learning models.

quant-ph↗

Fictitious Copy Quantum Error Mitigation

Errors are arguably the most pressing challenge impeding practical applications of quantum computers, which has instigated vigorous research on the development of quantum error mitigation (QEM) techniques. Existing QEM methods suppress errors with a varying degree of efficacy but importantly demand significant additional quantum and classical computational resources. In this work, we present Fictitious Copy Quantum Error Mitigation (FCQEM) method which corrects quantum errors without requiring any additional quantum resources and purely relies on using classical postprocessing of a joint probability distribution to correct expectation values. The joint probability distribution can be measured ``fictitiously'' by sampling one copy of noisy quantum circuit twice, or classically squaring probabilities from simply one copy. We show that FCQEM can recover eigenvalues even if exact eigenstates are not prepared. Furthermore, our technique can benefit other noise mitigation techniques with no additional quantum resources, which is demonstrated by combining FCQEM with the Quantum Computed Moments (QCM) method. FCQEM can compensate for noise that is pathological to QCM, and QCM allows for FCQEM to recover the ground state energy with a larger variety of trial states. We show that our technique can find the exact ground state energy of molecular and spin models under simulated noise models as well as experiments on a Rigetti 84-qubit superconducting quantum processor. The reported FCQEM method is general purpose for the current generation of quantum devices and is applicable to any problem that measures eigenvalues of operators on sharply peaked distributions.

quant-ph↗

Enhanced Random Subspace Local Projections for High-Dimensional Time Series Analysis

High-dimensional time series forecasting suffers from severe overfitting when the number of predictors exceeds available observations, making standard local projection methods unstable and unreliable. We propose an enhanced Random Subspace Local Projection (RSLP) framework designed to deliver robust impulse response estimation in the presence of hundreds of correlated predictors. The method introduces weighted subspace aggregation, category-aware subspace sampling, adaptive subspace size selection, and a bootstrap inference procedure tailored to dependent data. These enhancements substantially improve estimator stability at longer forecast horizons while providing more reliable finite-sample inference. Experiments on synthetic data, macroeconomic indicators, and the FRED-MD dataset demonstrate a 33 percent reduction in estimator variability at horizons h >= 3 through adaptive subspace size selection. The bootstrap inference procedure produces conservative confidence intervals that are 14 percent narrower at policy-relevant horizons in very high-dimensional settings (FRED-MD with 126 predictors) while maintaining proper coverage. The framework provides practitioners with a principled approach for incorporating rich information sets into impulse response analysis without the instability of traditional high-dimensional methods.

cs.LG↗

Quantum Jacobi-Davidson Method

Computing electronic structures of quantum systems is a key task underpinning many applications in photonics, solid-state physics, and quantum technologies. This task is typically performed through iterative algorithms to find the energy eigenstates of a Hamiltonian, which are usually computationally expensive and suffer from convergence issues. In this work, we develop and implement the Quantum Jacobi-Davidson (QJD) method and its quantum diagonalization variant, the Sample-Based Quantum Jacobi-Davidson (SBQJD) method, and demonstrate their fast convergence for ground state energy estimation. We assess the intrinsic algorithmic performance of our methods through exact numerical simulations on a variety of quantum systems, including 8-qubit diagonally dominant matrices, 12-qubit one-dimensional Ising models, and a 10-qubit water molecule (H$_2$O) Hamiltonian. Our results show that both QJD and SBQJD achieve significantly faster convergence and require fewer Pauli measurements compared to the recently reported Quantum Davidson method, with SBQJD further benefiting from optimized reference state preparation. These findings establish the QJD framework as an efficient general-purpose subspace-based technique for solving quantum eigenvalue problems, providing a promising foundation for sparse Hamiltonian calculations on future fault-tolerant quantum hardware.

quant-ph↗

Hybrid Action Reinforcement Learning for Quantum Architecture Search

Reinforcement learning-based Quantum Architecture Search (QAS) offers a promising avenue for automating the design of variational quantum circuits, but existing methods typically decouple discrete structure search from continuous parameter optimization, resulting in inefficient or brittle solutions. We propose HyRLQAS (Hybrid-Action Reinforcement Learning for Quantum Architecture Search), a unified reinforcement learning framework that jointly learns gate placement and parameter initialization within a hybrid discrete-continuous action space, while enabling dynamic refinement of previously placed gates. Trained in a variational quantum eigensolver setting, the agent constructs circuits that directly optimize molecular ground-state energies. Across multiple molecular benchmarks, HyRLQAS demonstrates strong and competitive performance against state-of-the-art QAS methods, achieving lower energy errors with fewer gates. Notably, HyRLQAS reaches chemical-accuracy-level convergence down to 1e-8 energy error after classical optimization, and policy-guided initialization reduces the iteration count of downstream classical optimizers. These results demonstrate that hybrid-action reinforcement learning provides a principled and effective mechanism for coupling circuit topology design with optimization-aware parameterization.

quant-ph↗

GCAMPS: A Scalable Classical Simulator for Qudit Systems

Classical simulations of quantum systems are notoriously difficult computational problems, with conventional state vector and tensor network methods restricted to quantum systems that feature only a small number of qudits. The recently introduced Clifford Augmented Matrix Product State (CAMPS) method offer scalability and efficiency by combining both tensor network and stabilizer simulation techniques and leveraging their complementary advantages. This hybrid simulation method has indeed demonstrated significant improvements in simulation performance for qubit circuits. Our work generalises the CAMPS method to higher quantum degrees of freedom -- qudit simulation, resulting in a generalised CAMPS (GCAMPS). Benchmarking this extended simulator on quantum systems with three degrees of freedom, i.e. qutrits, we show that similar to the case of qubits, qutrit systems also benefit from a comparable speedup using these techniques. Indeed, we see a greater improvement with qutrit simulation compared to qubit simulation on the same $T$-doped random Clifford benchmarking circuit as a result of the increased difficulty of conventional qutrit simulation using tensor networks. This extension allows for the classical simulation of problems that were previously intractable without access to a quantum device and will open new avenues to study complex many-body physics and to develop efficient methods for quantum information processing.

quant-ph↗

Adversarial Robustness Guarantees for Quantum Classifiers

Despite their ever more widespread deployment throughout society, machine learning algorithms remain critically vulnerable to being spoofed by subtle adversarial tampering with their input data. The prospect of near-term quantum computers being capable of running {quantum machine learning} (QML) algorithms has therefore generated intense interest in their adversarial vulnerability. Here we show that quantum properties of QML algorithms can confer fundamental protections against such attacks, in certain scenarios guaranteeing robustness against classically-armed adversaries. We leverage tools from many-body physics to identify the quantum sources of this protection. Our results offer a theoretical underpinning of recent evidence which suggest quantum advantages in the search for adversarial robustness. In particular, we prove that quantum classifiers are: (i) protected against weak perturbations of data drawn from the trained distribution, (ii) protected against local attacks if they are insufficiently scrambling, and (iii) show evidence that they are protected against universal adversarial attacks if they are sufficiently chaotic. Our analytic results are supported by numerical evidence demonstrating the applicability of our theorems and the resulting robustness of a quantum classifier in practice. This line of inquiry constitutes a concrete pathway to advantage in QML, orthogonal to the usually sought improvements in model speed or accuracy.

quant-ph↗

Efficient State Preparation for Quantum Machine Learning

One of the key considerations in the development of Quantum Machine Learning (QML) protocols is the encoding of classical data onto a quantum device. In this chapter we introduce the Matrix Product State representation of quantum systems and show how it may be used to construct circuits which encode a desired state. Putting this in the context of QML we show how this process may be modified to give a low depth approximate encoding and crucially that this encoding does not hinder classification accuracy and is indeed exhibits an increased robustness against classical adversarial attacks. This is illustrated by demonstrations of adversarially robust variational quantum classifiers for the MNIST and FMNIST dataset, as well as a small-scale experimental demonstration on a superconducting quantum device.

quant-ph↗

Time-Dynamic Circuits for Fault-Tolerant Shift Automorphisms in Quantum LDPC Codes

Quantum low-density parity-check (qLDPC) codes have emerged as a promising approach for realizing low-overhead logical quantum memories. Recent theoretical developments have established shift automorphisms as a fundamental building block for completing the universal set of logical gates for qLDPC codes. However, practical challenges remain because the existing SWAP-based shift automorphism yields logical error rates that are orders of magnitude higher than those for fault-tolerant idle operations. In this work, we address this issue by dynamically varying the syndrome measurement circuits to implement the shift automorphisms without reducing the circuit distance. We benchmark our approach on both twisted and untwisted weight-6 generalized toric codes, including the gross code family. Our time-dynamic circuits for shift automorphisms achieve performance comparable to the idle operations under the circuit-level noise model (SI1000). Specifically, the dynamic circuits achieve more than an order of magnitude reduction in logical error rates relative to the SWAP-based scheme for the gross code at a physical error rate of $10^{-3}$, employing the BP-OSD decoder. Our findings improve both the error resilience and the time overhead of the shift automorphisms in qLDPC codes. Furthermore, our work can lead to alternative syndrome extraction circuit designs, such as leakage removal protocols, providing a practical pathway to utilizing dynamic circuits that extend beyond surface codes towards qLDPC codes.

quant-ph↗

Quantum Error Correction and Detection for Quantum Machine Learning

At the intersection of quantum computing and machine learning, quantum machine learning (QML) is poised to revolutionize artificial intelligence. However, the vulnerability of the current generation of quantum computers to noise and computational error poses a significant barrier to this vision. Whilst quantum error correction (QEC) offers a promising solution for almost any type of hardware noise, its application requires millions of qubits to encode even a simple logical algorithm, rendering it impractical in the near term. In this chapter, we examine strategies for integrating QEC and quantum error detection (QED) into QML under realistic resource constraints. We first quantify the resource demands of fully error-corrected QML and propose a partial QEC approach that reduces overhead while enabling error correction. We then demonstrate the application of a simple QED method, evaluating its impact on QML performance and highlighting challenges we have yet to overcome before we achieve fully fault-tolerant QML.

quant-ph↗

Dual-Encoder Transformer-Based Multimodal Learning for Ischemic Stroke Lesion Segmentation Using Diffusion MRI

Accurate segmentation of ischemic stroke lesions from diffusion magnetic resonance imaging (MRI) is essential for clinical decision-making and outcome assessment. Diffusion-Weighted Imaging (DWI) and Apparent Diffusion Coefficient (ADC) scans provide complementary information on acute and sub-acute ischemic changes; however, automated lesion delineation remains challenging due to variability in lesion appearance. In this work, we study ischemic stroke lesion segmentation using multimodal diffusion MRI from the ISLES 2022 dataset. Several state-of-the-art convolutional and transformer-based architectures, including U-Net variants, Swin-UNet, and TransUNet, are benchmarked. Based on performance, a dual-encoder TransUNet architecture is proposed to learn modality-specific representations from DWI and ADC inputs. To incorporate spatial context, adjacent slice information is integrated using a three-slice input configuration. All models are trained under a unified framework and evaluated using the Dice Similarity Coefficient (DSC). Results show that transformer-based models outperform convolutional baselines, and the proposed dual-encoder TransUNet achieves the best performance, reaching a Dice score of 85.4% on the test set. The proposed framework offers a robust solution for automated ischemic stroke lesion segmentation from diffusion MRI.

eess.IV↗

PENDULUM: A Benchmark for Assessing Sycophancy in Multimodal Large Language Models

Sycophancy, an excessive tendency of AI models to agree with user input at the expense of factual accuracy or in contradiction of visual evidence, poses a critical and underexplored challenge for multimodal large language models (MLLMs). While prior studies have examined this behavior in text-only settings of large language models, existing research on visual or multimodal counterparts remains limited in scope and depth of analysis. To address this gap, we introduce a comprehensive evaluation benchmark, \textit{PENDULUM}, comprising approximately 2,000 human-curated Visual Question Answering pairs specifically designed to elicit sycophantic responses. The benchmark spans six distinct image domains of varying complexity, enabling a systematic investigation of how image type and inherent challenges influence sycophantic tendencies. Through extensive evaluation of state-of-the-art MLLMs. we observe substantial variability in model robustness and a pronounced susceptibility to sycophantic and hallucinatory behavior. Furthermore, we propose novel metrics to quantify sycophancy in visual reasoning, offering deeper insights into its manifestations across different multimodal contexts. Our findings highlight the urgent need for developing sycophancy-resilient architectures and training strategies to enhance factual consistency and reliability in future MLLMs. Our proposed dataset with MLLMs response are available at https://github.com/ashikiut/pendulum/.

cs.AI↗

Complex Swin Transformer for Accelerating Enhanced SMWI Reconstruction

Susceptibility Map Weighted Imaging (SMWI) is an advanced magnetic resonance imaging technique used to detect nigral hyperintensity in Parkinsons disease. However, full resolution SMWI acquisition is limited by long scan times. Efficient reconstruction methods are therefore required to generate high quality SMWI from reduced k space data while preserving diagnostic relevance. In this work, we propose a complex valued Swin Transformer based network for super resolution reconstruction of multi echo MRI data. The proposed method reconstructs high quality SMWI images from low resolution k space inputs. Experimental results demonstrate that the method achieves a structural similarity index of 0.9116 and a mean squared error of 0.076 when reconstructing SMWI from 256 by 256 k space data, while maintaining critical diagnostic features. This approach enables high quality SMWI reconstruction from reduced k space sampling, leading to shorter scan times without compromising diagnostic detail. The proposed method has the potential to improve the clinical applicability of SMWI for Parkinsons disease and support faster and more efficient neuroimaging workflows.

eess.IV↗

DGTEN: A Robust Deep Gaussian based Graph Neural Network for Dynamic Trust Evaluation with Uncertainty-Quantification Support

Dynamic trust evaluation in large, rapidly evolving graphs demands models that capture changing relationships, express calibrated confidence, and resist adversarial manipulation. DGTEN (Deep Gaussian-Based Trust Evaluation Network) introduces a unified graph-based framework that does all three by combining uncertainty-aware message passing, expressive temporal modeling, and built-in defenses against trust-targeted attacks. It represents nodes and edges as Gaussian distributions so that both semantic signals and epistemic uncertainty propagate through the graph neural network, enabling risk-aware trust decisions rather than overconfident guesses. To track how trust evolves, it layers hybrid absolute-Gaussian-hourglass positional encoding with Kolmogorov-Arnold network-based unbiased multi-head attention, then applies an ordinary differential equation-based residual learning module to jointly model abrupt shifts and smooth trends. Robust adaptive ensemble coefficient analysis prunes or down-weights suspicious interactions using complementary cosine and Jaccard similarity, curbing reputation laundering, sabotage, and on-off attacks. On two signed Bitcoin trust networks, DGTEN delivers standout gains where it matters most: in single-timeslot prediction on Bitcoin-OTC, it improves MCC by +12.34% over the best dynamic baseline; in the cold-start scenario on Bitcoin-Alpha, it achieves a +25.00% MCC improvement, the largest across all tasks and datasets; while under adversarial on-off attacks, it surpasses the baseline by up to +10.23% MCC. These results endorse the unified DGTEN framework.

cs.LG↗

Anatomy-Guided Representation Learning Using a Transformer-Based Network for Thyroid Nodule Segmentation in Ultrasound Images

Accurate thyroid nodule segmentation in ultrasound images is critical for diagnosis and treatment planning. However, ambiguous boundaries between nodules and surrounding tissues, size variations, and the scarcity of annotated ultrasound data pose significant challenges for automated segmentation. Existing deep learning models struggle to incorporate contextual information from the thyroid gland and generalize effectively across diverse cases. To address these challenges, we propose SSMT-Net, a Semi-Supervised Multi-Task Transformer-based Network that leverages unlabeled data to enhance Transformer-centric encoder feature extraction capability in an initial unsupervised phase. In the supervised phase, the model jointly optimizes nodule segmentation, gland segmentation, and nodule size estimation, integrating both local and global contextual features. Extensive evaluations on the TN3K and DDTI datasets demonstrate that SSMT-Net outperforms state-of-the-art methods, with higher accuracy and robustness, indicating its potential for real-world clinical applications.

cs.CV↗

Integrating Uncertainty Quantification into Computational Fluid Dynamics Models of Coronary Arteries Under Steady Flow

Computational models are continuously integrated in the clinical space, where they support clinicians in disease diagnosis, prognosis, and prevention strategies. While assisting in clinical space, these computational models frequently use deterministic approaches, where the inherent (aleatoric) variability of input parameters is ignored. This questions the credibility and often hinders the clinical adoption of these computational models. Therefore, in this study, we introduced uncertainty quantification in the computational fluid dynamics models of the left main coronary artery to analyze the influence of input hemodynamics parameters on wall shear stress (WSS). UncertainSCI was used, where an emulator was built using polynomial chaos expansion between the input parameters and the output quantity of interest, and the output sensitivities and statistics were directly extracted from the emulator. The uncertainty-informed framework was first applied to an analytical solution of the Navier-Stokes equation (Poiseuille flow) and then to a patient-specific model of the left main coronary artery. Different input hemodynamics parameters are considered, such as pressure, viscosity, density, velocity, and radius, whereas wall shear stress was considered as our output quantity of interest. The results suggest that velocity dominated the variability in WSS in the analytical model (~79%), whereas viscosity dominated in the patient-specific model (~59%). The results further suggest that out of all the Sobol indices interactions, unary interactions were the most dominant ones, contributing ~93.2% and ~99% for the analytical and patient-specific model, respectively. This study will enhance confidence in computational models, facilitating their adoption in the clinical space to improve decision-making for coronary artery disease diagnosis, prognosis, and therapeutic strategies.

physics.med-ph↗