arXiv subjects

Thai T. Vu

Publications and source records attributed to Thai T. Vu.

10 recordsLinked to original sources

arXiv2026

Proportional-Fair Resource Allocation and Dual-Threshold Early-Exit Inference for Secure Cooperative Multi-Layer Edge Intelligence

This paper proposes FREDI (Fair Resource Allocation for Edge Dual-Threshold Inference), a secure wireless edge-intelligence framework for event-triggered inference in a cooperative user equipment (UE)--edge server (ES)--cloud system. Each UE performs early-exit convolutional neural network (CNN) screening using dual confidence thresholds, while critical events are securely offloaded to an edge server for detailed classification. We formulate a proportionally-fair utility maximization problem that jointly optimizes UE--ES association, wireless and processing resources, and confidence thresholds. FREDI decomposes the problem into proportional-fair resource allocation and dual-threshold inference optimization. We prove that the detected-critical event set is set-monotone non-increasing in both thresholds, and exploit the finite empirical confidence domain for exact threshold optimization. An empirical resource--utility response envelope yields a computable global suboptimality bound and a sufficient condition for global optimality. By pre-eliminating infeasible UE--ES pairs and exactly projecting out bandwidth and transmit-power variables, the resource-allocation subproblem is reduced to a mixed-integer exponential-cone program solvable to the certified global optimality within a prescribed gap. Numerical results with early-exit MobileNetV2 and ShuffleNetV2 demonstrate near-perfect UE fairness with aggregate utility close to a Sum-Utility benchmark, reveal security-induced resource fragmentation, and demonstrate the Stage-A scalability from 6 to 144 UEs with median solving time below 0.1~s in the tested configurations.

cs.NI↗

arXiv2026

MetaMorphQ: Physics-Based Metamorphic Testing of Variational Quantum Circuits

Variational Quantum Eigensolvers (VQEs) are central to quantum computing, yet testing them remains challenging due to the oracle problem: the ground-state energy they compute is itself unknown. Existing approaches, such as convergence-based testing, are unreliable and yield high false-positive rates due to optimisation instability. We propose METAMORPHQ, a metamorphic testing framework that derives test oracles directly from quantum mechanical properties of VQE circuits. Exploiting algebraic properties of parametrised rotation gates and diagonal Hamiltonians, we define five physics-based invariants that hold for any correct circuit and can be verified at initialisation without ground-truth outputs. Evaluated on 500 benchmark circuits with 2,469 mutants, METAMORPHQ achieves zero false positives and significantly improves diagnostic effectiveness (Youden's J = 0.57 vs. 0.02 for convergence testing). These results demonstrate that physics-derived invariants provide a practical, oracle-free foundation for testing quantum software, enabling reliable validation of both human- and LLM-generated circuits.

cs.ET↗

arXiv2026

TGHE: Template-based Graph Homomorphic Encryption for Privacy-Preserving GNN Inference in Edge-Cloud Systems

Existing homomorphic encryption (HE)-based GNN systems adopt a graph-centric paradigm that couples per-query cost to global graph size, limiting evaluations to at most ~20k nodes and making them incompatible with dynamic, large-scale financial graphs. We propose TGHE (Template-based Graph Homomorphic Encryption), an ego-centric framework that resolves this by exploiting a template phenomenon: local computation trees in transaction graphs converge into a small set of structural shapes. TGHE canonicalizes ego-graphs at the edge and packs structurally identical trees into shared CKKS ciphertexts for SIMD-parallel encrypted inference, with two long-tail optimizers (Approximate Template Fitting and Topology Collapse) ensuring full SIMD coverage. On DGraphFin (3.7M nodes, 4.3M edges), TGHE-Collapse achieves a 66.9x speedup over the sequential encrypted baseline with less than 0.002 AUC loss.

cs.CR↗

arXiv2026

MPE-Adam: Multi-Population Evolutionary Optimization with Adam Refinement for QAOA

Parameter optimization is a central bottleneck in variational quantum algorithms such as the Quantum Approximate Optimization Algorithm (QAOA). The classical optimizer must navigate a high-dimensional, non-convex parameter space under measurement noise. From a quantum software perspective, this process forms a multi-stage workflow: global exploration of the parameter space followed by local refinement within the hybrid quantum-classical loop. Most existing approaches, however, employ single-stage optimizers that do not separate these roles, which limits the use of complementary strategies. We propose MPE-Adam, a hybrid optimization framework that integrates multi-population evolutionary search for global exploration with Adam-based gradient refinement for local convergence. The method is structured as a modular component suitable for quantum software pipelines. We evaluate MPE-Adam on MaxCut instances generated from random 3-regular graphs with up to 22 nodes. The results show that MPE-Adam achieves higher approximation ratios and lower variance than evolutionary-only and SPSA-based baselines, with statistically significant improvements. These findings indicate that structured multi-stage optimization improves both solution quality and software-level flexibility in quantum applications.

cs.ET↗

arXiv2025

RAID: Refusal-Aware and Integrated Decoding for Jailbreaking LLMs

Large language models (LLMs) achieve impressive performance across diverse tasks yet remain vulnerable to jailbreak attacks that bypass safety mechanisms. We present RAID (Refusal-Aware and Integrated Decoding), a framework that systematically probes these weaknesses by crafting adversarial suffixes that induce restricted content while preserving fluency. RAID relaxes discrete tokens into continuous embeddings and optimizes them with a joint objective that (i) encourages restricted responses, (ii) incorporates a refusal-aware regularizer to steer activations away from refusal directions in embedding space, and (iii) applies a coherence term to maintain semantic plausibility and non-redundancy. After optimization, a critic-guided decoding procedure maps embeddings back to tokens by balancing embedding affinity with language-model likelihood. This integration yields suffixes that are both effective in bypassing defenses and natural in form. Experiments on multiple open-source LLMs show that RAID achieves higher attack success rates with fewer queries and lower computational cost than recent white-box and black-box baselines. These findings highlight the importance of embedding-space regularization for understanding and mitigating LLM jailbreak vulnerabilities.

cs.CL↗

arXiv2025

Quantum Machine Learning for Secure Cooperative Multi-Layer Edge AI with Proportional Fairness

This paper proposes a communication-efficient, event-triggered inference framework for cooperative edge AI systems comprising multiple user devices and edge servers. Building upon dual-threshold early-exit strategies for rare-event detection, the proposed approach extends classical single-device inference to a distributed, multi-device setting while incorporating proportional fairness constraints across users. A joint optimization framework is formulated to maximize classification utility under communication, energy, and fairness constraints. To solve the resulting problem efficiently, we exploit the monotonicity of the utility function with respect to the confidence thresholds and apply alternating optimization with Benders decomposition. Experimental results show that the proposed framework significantly enhances system-wide performance and fairness in resource allocation compared to single-device baselines.

cs.NI↗

arXiv2023

Energy-based Proportional Fairness in Cooperative Edge Computing

By executing offloaded tasks from mobile users, edge computing augments mobile user equipments (UEs) with computing/communications resources from edge nodes (ENs), enabling new services (e.g., real-time gaming). However, despite being more resourceful than UEs, allocating ENs' resources to a given favorable set of users (e.g., closer to ENs) may block other UEs from their services. This is often the case for most existing approaches that only aim to maximize the network social welfare or minimize the total energy consumption but do not consider the computing/battery status of each UE. This work develops an energy-based proportional-fair framework to serve all users with multiple tasks while considering both their service requirements and energy/battery levels in a multi-layer edge network. The resulting problem for offloading tasks and allocating resources toward the tasks is a Mixed-Integer Nonlinear Programming, which is NP-hard. To tackle it, we leverage the fact that the relaxed problem is convex and propose a distributed algorithm, namely the dynamic branch-and-bound Benders decomposition (DBBD). DBBD decomposes the original problem into a master problem (MP) for the offloading decisions and multiple subproblems (SPs) for resource allocation. To quickly eliminate inefficient offloading solutions, MP is integrated with powerful Benders cuts exploiting the ENs' resource constraints. We then develop a dynamic branch-and-bound algorithm (DBB) to efficiently solve MP considering the load balance among ENs. SPs can either be solved for their closed-form solutions or be solved in parallel at ENs, thus reducing the complexity. The numerical results show that DBBD returns the optimal solution in maximizing the proportional fairness among UEs. DBBD has higher fairness indexes, i.e., Jain's index and min-max ratio, in comparison with the existing ones that minimize the total consumed energy.

cs.NI↗

arXiv2020

Optimal Energy Efficiency with Delay Constraints for Multi-layer Cooperative Fog Computing Networks

We develop a joint offloading and resource allocation framework for a multi-layer cooperative fog computing network, aiming to minimize the total energy consumption of multiple mobile devices subject to their service delay requirements. The resulting optimization involves both binary (offloading decisions) and real variables (resource allocations), making it an NP-hard and computationally intractable problem. To tackle it, we first propose an improved branch-and-bound algorithm (IBBA) that is implemented in a centralized manner. However, due to the large size of the cooperative fog computing network, the computational complexity of the proposed IBBA is relatively high. To speed up the optimal solution searching as well as to enable its distributed implementation, we then leverage the unique structure of the underlying problem and the parallel processing at fog nodes. To that end, we propose a distributed framework, namely feasibility finding Benders decomposition (FFBD), that decomposes the original problem into a master problem for the offloading decision and subproblems for resource allocation. The master problem (MP) is then equipped with powerful cutting-planes to exploit the fact of resource limitation at fog nodes. The subproblems (SP) for resource allocation can find their closed-form solutions using our fast solution detection method. These (simpler) subproblems can then be solved in parallel at fog nodes. The numerical results show that the FFBD always returns the optimal solution of the problem with significantly less computation time (e.g., compared with the centralized IBBA approach). The FFBD with the fast solution detection method, namely FFBD-F, can reduce up to $60\%$ and $90\%$ of computation time, respectively, compared with those of the conventional FFBD, namely FFBD-S, and IBBA.

cs.NI↗

arXiv2018

Offloading Energy Efficiency with Delay Constraint for Cooperative Mobile Edge Computing Networks

We propose a novel edge computing network architecture that enables edge nodes to cooperate in sharing computing and radio resources to minimize the total energy consumption of mobile users while meeting their delay requirements. To find the optimal task offloading decisions for mobile users, we first formulate the joint task offloading and resource allocation optimization problem as a mixed integer non-linear programming (MINLP). The optimization involves both binary (offloading decisions) and real variables (resource allocations), making it an NP-hard and computational intractable problem. To circumvent, we relax the binary decision variables to transform the MINLP to a relaxed optimization problem with real variables. After proving that the relaxed problem is a convex one, we propose two solutions namely ROP and IBBA. ROP is adopted from the interior point method and IBBA is developed from the branch and bound algorithm. Through the numerical results, we show that our proposed approaches allow minimizing the total energy consumption and meet all delay requirements for mobile users.

cs.NI↗

arXiv2018

2D Proactive Uplink Resource Allocation Algorithm for Event Based MTC Applications

We propose a two dimension (2D) proactive uplink resource allocation (2D-PURA) algorithm that aims to reduce the delay/latency in event-based machine-type communications (MTC) applications. Specifically, when an event of interest occurs at a device, it tends to spread to the neighboring devices. Consequently, when a device has data to send to the base station (BS), its neighbors later are highly likely to transmit. Thus, we propose to cluster devices in the neighborhood around the event, also referred to as the disturbance region, into rings based on the distance from the original event. To reduce the uplink latency, we then proactively allocate resources for these rings. To evaluate the proposed algorithm, we analytically derive the mean uplink delay, the proportion of resource conservation due to successful allocations, and the proportion of uplink resource wastage due to unsuccessful allocations for 2D-PURA algorithm. Numerical results demonstrate that the proposed method can save over 16.5 and 27 percent of mean uplink delay, compared with the 1D algorithm and the standard method, respectively.

cs.NI↗