Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Muhammad Waqas Arshad has a total of 64 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Leakage-Controlled Projected Quantum Kernels for Orbital Conjunction Risk Assessment Using SOCRATES Data

    by Muhammad Waqas Arshad, David Q. Liu, Syed Rizwan Hassan

    Conjunction assessment needs risk screening that is early, scalable and interpretable. We ask whether projected quantum feature maps add predictive value to this task when compared with strong classical learners under strict leakage control. From 149,044 real SOCRATES conjunction events (119,568 object pairs), split pair-disjointly, we predict the base-ten logarithm of the maximum collision probability. We prove that the pair-disjoint held-out loss is unbiased for the risk on unseen pairs, that the logarithmic target preserves risk ranking, that the local six-qubit feature map is exactly a classical additive model, and when a validation-fitted ensemble assigns a member zero weight. In the configuration fixed a priori (encoding bandwidth π, linear outer kernel), the projected models trail gradient boosting (RMSE 0.668 and 0.762 against 0.602), and the local map is no better than a linear ridge on the raw inputs. Matched experiments on identical training sets show that this shortfall comes from two design choices rather than from the representation: the bandwidth folds the dominant descriptors, and the linear outer kernel discards nonlinear structure. With a validated bandwidth and a Gaussian outer kernel, the local projected model outperforms a tuned classical Gaussian kernel on 1200-event subsets in all five draws (0.641±0.003) and has the lowest point estimate at full scale (0.589), where it is statistically tied with that kernel. It takes the largest weight in an ensemble that lowers RMSE from 0.599 to 0.579. This is not a quantum advantage: the local model is classically computable in closed form, and the interaction pattern of the entangled variant never improves accuracy. In 200 randomly seeded configurations at B=π and B=π/4, it reduces it, and at the validated bandwidth the difference is not significant at full scale.

  • Open Access

    CAGE-QMol: A Constraint-Aware Quantum-Inspired Optimization Framework for Brain-Penetrant Multi-Target Alzheimer’s Drug Discovery

    by Muhammad Waqas Arshad, David Q. Liu, Muhammad Bilal Sarwar, Syed Rizwan Hassan, KangYoon Lee

    Alzheimer’s disease remains one of the hardest disorders to drug, and most computational pipelines still tackle one target at a time and ignore brain penetration until late. We bring all of that into a single optimization problem. Our framework, called CAGE-QMol(Constraint-Aware Quantum-inspired Molecular optimization), turns the search for a brain-penetrant, dual BACE1/AChE inhibitor into a constrained quadratic unconstrained binary optimization (QUBO) and solves it with an ensemble of classical, quantum-inspired, and Quantum Approximate Optimization Algorithm (QAOA) backends. The pipeline begins with three real public datasets—MoleculeNet BACE, ChEMBL CHEMBL4822 (BACE1) and CHEMBL220 (AChE), and the TDC BBB_Martins blood–brain-barrier set—which together yield 12,465 unique molecules with at least one measured endpoint. Multi-task ExtraTrees predictors trained on Morgan ECFP4 fingerprints and physicochemical descriptors deliver scaffold-split test-set metrics of R2=0.624 (MAE=0.587) for BACE1 and R2=0.383 (MAE=0.793) for AChE. The optimizer combines these predictions with a Lipinski-based feasibility cone, a TDC-derived BBB classifier, and similarity-driven diversity into a constrained QUBO. We adapt the classical exact-penalty rule, λ>ΔS/δg, which guarantees every global minimizer of the penalized energy is feasible, and specialize it so that the multiplier is computed from the data rather than hand-tuned. A 10-qubit PennyLane QAOA circuit is benchmarked against exact enumeration, simulated annealing, genetic search, Bayesian TPE, and random search across ten seeds; the QUBO formulation lets a genetic solver match the exact ground state on every seed, while ablations show that removing the QUBO selection collapses the mean therapeutic score from 6.242 to 5.976 (p<10−3, paired t-test). Top-50 candidates exhibit a mean BBB probability of 0.716, a mean QED of 0.768, and 100% Lipinski feasibility, with leading scaffolds (tetrahydroisoquinolinone, methoxy-tetrahydronaphthalene-urea, indanone-piperidine) reproducing motifs found in published dual BACE1/AChE inhibitor families. This paper contributes (i) a mathematically grounded penalty selection rule for constrained drug-discovery QUBOs, (ii) a single end-to-end pipeline from raw public data to ranked, 3D-embedded leads, and (iii) reproducible head-to-head benchmarks between classical, quantum-inspired, and QAOA-simulated optimizers on a real Alzheimer’s task.

  • Open Access

    Calibration-aware gating for budgeted LLM supervision in graph neural network link prediction

    by Muhammad Waqas Arshad, Stefano Lodi, David Q. Liu, Usman Adeel, Syed Rizwan Hassan

    Large-scale graphs underpin critical applications such as financial fraud detection, citation analysis, and biomedical discovery, where link prediction is often performed under severe label scarcity, class imbalance, and operational cost constraints. Recent large language models (LLMs) provide powerful semantic priors that can augment graph learning; however, naïvely querying LLMs is computationally expensive, and direct distillation of their outputs into compact graph models can degrade predictive reliability. In this work, we propose a budgeted LLM-assisted framework for graph link prediction that explicitly accounts for both predictive performance and calibration. The framework consists of four stages: (i) training a backbone graph neural network (GNN) using available labeled edges, (ii) selectively querying an LLM under a fixed query budget, (iii) distilling LLM-provided soft supervision into a lightweight student predictor, and (iv) learning a calibration-aware gating model that adaptively combines base and student predictions on a per-edge basis. The proposed gate relies only on inexpensive uncertainty- and structure-aware signals, such as predictive entropy and node degree, to determine when LLM supervision is likely to be beneficial. We evaluate the proposed approach using both ranking and reliability metrics, including ROC-AUC, Average Precision (AP), Expected Calibration Error (ECE), and Brier score. Experiments on four benchmark datasets—Elliptic, Cora, PubMed, and OGBN-Arxiv—demonstrate that the learned gate consistently improves ranking performance over both the base GNN and the LLM-distilled student under the same query budget, while substantially reducing calibration error. These results highlight that selective, calibration-aware LLM supervision offers a practical and trustworthy pathway for improving graph link prediction in cost-sensitive real-world settings.

  • Open Access

    HPC: A Computational Benchmark of Classical, Parallel, and Hybrid Metaheuristics for QUBO-Based Suspension Geometry Optimization

    by Muhammad Waqas Arshad, Stefano Lodi, Omair Ashraf, Muhammad Haseeb Rasool, Syed Rizwan Hassan

    The calibration of suspension geometry involves highly nonlinear kinematic relationships and leads to challenging optimization landscapes that are difficult to explore efficiently with classical local methods. Quadratic Unconstrained Binary Optimization (QUBO) provides a unified discrete formulation that enables the use of a wide range of metaheuristic solvers, but its practical behavior in realistic engineering-inspired problems remains insufficiently benchmarked. This paper presents a computational benchmarking study of classical, parallel, and hybrid metaheuristic solvers applied to a QUBO-formulated double wishbone suspension geometry problem. A symbolic multi-body kinematic model is constructed and discretized into a large-scale QUBO instance capturing camber and caster tracking objectives across multiple roll conditions. Using a fixed low-resolution binary encoding, we systematically evaluate solver performance in terms of objective value, runtime, and time-to-solution trade-offs. The benchmark includes standard simulated annealing and tabu search, parallel simulated annealing, population-based annealing, bandit-controlled hybrid heuristics, and continuous-relaxation-based ADMM methods with and without annealing refinement. Extensive experiments conducted on a Euro-HPC pre-exascale system demonstrate that parallel and hybrid solvers can achieve substantial reductions in wall-clock time—often exceeding two orders of magnitude—while attaining objective values comparable to classical simulated annealing. The results reveal clear trade-offs between solution quality and computational efficiency, and highlight how solver structure influences performance on large QUBO instances derived from symbolic engineering models. Rather than focusing on final design optimality, this study provides a reproducible reference benchmark and practical insights into solver behavior for QUBO-based engineering optimization problems.

  • Open Access

    An Advanced Hybrid Optimization Algorithm for Vehicle Suspension Design Using a QUBO-SQP Framework

    by Muhammad Waqas Arshad, Stefano Lodi, David Q. Liu

    The design of multi-link vehicle suspension systems, such as the 3D double-wishbone, presents a critical challenge in automotive engineering. The process constitutes a high-dimensional, nonlinearly constrained optimization problem where traditional gradient-based methods often fail by converging to suboptimal local minima. This paper introduces a novel two-stage hybrid optimization framework designed to overcome this limitation by intelligently integrating quantum-inspired and classical techniques. The methodology explicitly defines a QUBO (Quadratic Unconstrained Binary Optimization) stage and an SQP (Sequential Quadratic Programming) stage. Stage 1 addresses the complex kinematic constraint problem by formulating it as a QUBO, which is then solved using Simulated Annealing to perform a global search, guaranteeing a physically feasible starting point. Subsequently, Stage 2 takes this feasible solution and employs an SQP algorithm to perform a high-precision local refinement, tuning the geometry to meet specific performance targets for camber and caster curves. The framework successfully converged to a design with a near-zero performance objective of 7.08 × 10−14. The efficacy of this hybrid approach is highlighted by the dramatic improvement from the high-error initial solution found by Simulated Annealing to the final, high-precision result from the SQP refinement. We conclude that this QUBO-SQP framework is a powerful and validated methodology for solving complex, real-world engineering design problems, effectively bridging the gap between global exploration and local precision.


  • Open Access

    Double Wishbone Suspension: A Computational Framework for Parametric 3D Kinematic Modeling and Simulation Using Mathematica

    by Muhammad Waqas Arshad, Stefano Lodi, David Q. Liu

    The double wishbone suspension (DWS) system is widely used in automotive engineering because of its favorable kinematic properties, which affect vehicle dynamics, handling, and ride comfort; hence, it is important to have an accurate 3D model, simulation, and analysis of the system in order to optimize its design. This requires efficient computational tools for parametric study. The development of effective computational tools that support parametric exploration stands as an essential requirement. Our research demonstrates a complete Wolfram Mathematica system that creates parametric 3D kinematic models and conducts simulations, performs analyses, and generates interactive visualizations of DWS systems. The system uses homogeneous transformation matrices to establish the spatial relationships between components relative to a global coordinate system. The symbolic geometric parameters allow designers to perform flexible design exploration and the kinematic constraints create an algebraic equation system. The numerical solution function NSolve computes linkage positions from input data, which enables fast evaluation of different design parameters. The integrated 3D visualization module based on Mathematica’s manipulate function enables users to see immediate results of geometric configurations and parameter effects while calculating exact 3D coordinates. The resulting robust, systematic, and flexible computational environment integrates parametric 3D design, kinematic simulation, analysis, and dynamic visualization for DWS, serving as a valuable and efficient tool for engineers during the design, development, assessment, and optimization phases of these complex automotive systems.


  • Open Access

    A Hybrid Genetic Algorithm and Hidden Markov Model-Based Hashing Technique for Robust Data Security

    by King Faisal University, Aseel Alshuaibi, Muhammad Waqas Arshad, University of Bologna, Mohammed Maayah, INTI International University

    The growing dependence on technology to store, process, and transmit data across interconnected systems has significantly elevated the need for robust data security. Modern computer systems emphasize the critical principles of authentication and data integrity. With the rise in cyber threats, the importance of securing data transactions against unauthorized or unintentional modifications has become more apparent than ever. As computers continue to play an increasingly vital role in daily operations, managing and safeguarding data alterations is essential. To address these challenges, businesses must adopt proactive measures to reinforce the security of sensitive data and passwords. Hashing functions, a well-established cryptographic approach, have proven effective in addressing a wide range of authentication and data integrity issues. A hash function generates a fixed-length output, known as a "digest," from an input. This one-way function is irreversible, providing a secure method of encoding data. However, hash functions are still vulnerable to various attacks, including dictionary attacks, brute-force attacks, and the use of lookup tables. The strength of a hashing function can be evaluated based on the number of attempts required to break it, the size of the hash key, and the specific algorithm employed. In response, this study proposes a novel hashing technique that integrates a Genetic Algorithm (GA) and Hidden Markov Models within a block hashing framework. Inspired by evolutionary biology, the GA applies operators such as mutation, crossover (recombination), and selection to simulate natural selection, offering a dynamic and efficient method for enhancing data security. The proposed algorithm utilizes the Hill cipher as an encryption mechanism and incorporates a singular Hill cipher key matrix to enhance security and reduce the likelihood of hash collisions or reversals. Experimental results demonstrate that the developed algorithm exhibits strong resistance to multiple attack types and outperforms several existing methods in terms of accuracy and robustness.


  • Cutting-Edge Malware Detection in Healthcare: Leveraging Cascaded-AlexNet Model

    by Sania Akhtar, Muhammad Hanif, Muhammad Waqas Arshad, Faryal Farooq


  • YOLOv11-SAMNet: A Hybrid Detection and Segmentation Framework for Urine Sediment Analysis

    by Sania Akhtar, Muhammad Hanif, Hamdi Melih Saraoglu, Sham Lal, Muhammad Waqas Arshad