arXiv subjects

Kien X. Nguyen

Publications and source records attributed to Kien X. Nguyen.

12 recordsLinked to original sources

arXiv2026

Scaling Quantum Optimization for Unit Commitment via Pauli Correlation Encoding

Unit commitment is an important optimization problem in power system operations, classified as NP-hard. This paper presents a hybrid quantum-classical method for the unit commitment problem with time-dependent constraints, where decisions must be made about which generators to turn on/off and how much power they should produce over a planning horizon. We use a hybrid quantum-classical optimization procedure to determine the on/off schedules of the generating units and the corresponding power dispatch that satisfies operational constraints such as load balance, generator limits, ramping, and reserve requirements. We frame the optimization loop as a leader-follower structure, where the quantum optimizer leads to give the on/off decisions, and the classical optimizer follows to produce the power level schedule. Leveraging Pauli-Correlation Encoding, our method scales to horizon-wide unit commitment schedules by encoding the binary variables with far fewer qubits. By combining these components, the method can handle multi-period settings while using far fewer qubits than straightforward quantum encodings that allocate one qubit per decision variable as in prior approaches. We evaluate the approach on both small- and large-scale instances, up to 312 binary variables, and show that it reliably produces feasible schedules with competitive operating costs.

quant-ph↗

arXiv2026

QAP-Router: Tackling Qubit Routing as Dynamic Quadratic Assignment with Reinforcement Learning

Qubit routing is a fundamental problem in quantum compilation, known to be NP-hard. Its dynamic nature makes local routing decisions propagate and compound over time, making global efficient solutions challenging. Existing heuristic methods rely on local rules with limited lookahead, while recent learning-based approaches often treat routing as a generic sequential decision problem without fully exploiting its underlying structure. In this paper, we introduce QAP-Router, framing qubit routing based on a dynamic Quadratic Assignment Problem (QAP) formulation. By modeling logical interactions, or quantum gates, as flow matrices and hardware topology as a distance matrix, our approach captures the interaction-distance coupling in a unified objective, which defines the reward in the reinforcement learning environment. To further exploit this structure, the policy network employs a solution-aware Transformer backbone that encodes the interaction between the flow matrix and the distance matrix into the attention mechanism. We also integrate a lookahead mechanism that blends naturally into the QAP framework, preventing myopic decisions. Extensive experiments on 1,831 real-world quantum circuits from the MQTBench, AgentQ and QUEKO datasets show that our method substantially reduces the CNOT gate count of routed circuits by 15.7%, 30.4% and 12.1%, respectively, relative to existing industry compilers.

quant-ph↗

arXiv2026

Quantum Hypergraph Partitioning

Quantum optimization algorithms are inherently probabilistic, yet they are most often used to search for a single high-quality solution. In this paper, we instead study hypergraph partitioning problems in which the desired output is itself a probability distribution over partitions. We introduce a distributional perspective on hypergraph partitioning motivated by maximin and minimax objectives such as Fair Cut Cover, and we show how these objectives align naturally with the measurement distribution produced by QAOA. To motivate the formulation, we introduce a workforce-scheduling-inspired toy problem, the Greatest Expected Imbalance problem, in which the goal is to minimize the worst expected imbalance across hyperedges. We then develop QAOA-based quantum solvers that represent distributional solutions natively through quantum states, together with quadratic hypergraph objectives suitable for standard and multi-objective QAOA. These formulations connect balanced hypergraph partitioning, polarized community discovery, and distributional fairness under a unified quantum optimization framework. For comparison, we provide optimal polynomial-time classical approximation algorithms based on semidefinite programming and hyperplane rounding. Experiments on real-world and synthetic hypergraphs demonstrate that low-depth multi-angle QAOA can outperform these classical approximation baselines on the proposed objectives, highlighting the potential of quantum algorithms for optimization problems where the solution is a distribution rather than a single partition.

quant-ph↗

arXiv2026

Q3SAT-GPT: A Generative Model for Discovering Quantum Circuits for the 3-SAT Problem

This work introduces Q3SAT-GPT, a generative model for discovering quantum circuits for the Max-E3-SAT problem. Our method learns from high-performing QAOA-style ansätze to directly generate candidate circuits. To create high-quality supervision, we also introduce Mosaic Adaptive QAOA (MosaicADAPT-QAOA), an adaptive strategy for constructing low-depth QAOA circuits by selecting subsets of mixer operators in each step, rather than inserting operators sequentially. The resulting circuits serve as training data for the generative model, allowing it to learn effective circuit design patterns while eliminating the need for costly variational optimization at inference time. Experiments show that our framework attains strong solution quality with shallow circuits and scales significantly better than both our adaptive construction procedure and conventional variational baselines. Our results establish generative modeling as a high-performance route toward the scalable discovery of quantum optimization circuits, demonstrating that these models can effectively internalize circuit logic while providing a foundation for future, instance-aware inductive biases. Reproducibility: The source code is available at https://github.com/pratimugale/Q3SAT-GPT.

quant-ph↗

arXiv2026

Graph-Conditioned Meta-Optimizer for QAOA Parameter Generation on Multiple Problem Classes

We study parameter transferability for the Quantum Approximate Optimization Algorithm (QAOA) across multiple combinatorial optimization problem classes from a parameter generation perspective. Specifically, a meta-optimizer is trained on one problem class and deployed on another during test time. Prior work employs a Long Short-Term Memory network to emulate QAOA optimization trajectories, but the learned dynamics usually collapse to near-identical paths, limiting cross-problem transfer efficiency. In this paper, we present a problem-aware graph-conditioned meta-optimizer for QAOA that learns to generate parameter trajectories over a fixed horizon, providing strong initializations with only a few steps. The optimizer is conditioned on compact graph embeddings and trained end-to-end using differentiable feedback from the QAOA objective, avoiding the need for ground-truth angles. We evaluate across multiple graph problem classes, including MaxCut, Maximum Independent Set, Maximum Clique, and Minimum Vertex Cover. We report both solution quality and feasibility-aware metrics where constraints apply. Results across a comprehensive empirical study consisting of 64 settings show that the learned optimizer can reduce optimization effort and improve performance over standard initialization, while exhibiting transferable behavior across graph families and problem types.

quant-ph↗

arXiv2026

UniHetCO: A Unified Heterogeneous Representation for Multi-Problem Learning in Unsupervised Neural Combinatorial Optimization

Unsupervised neural combinatorial optimization (NCO) offers an appealing alternative to supervised approaches by training learning-based solvers without ground-truth solutions, directly minimizing instance objectives and constraint violations. Yet for graph node subset-selection problems (e.g., Maximum Clique and Maximum Independent Set), existing unsupervised methods are typically specialized to a single problem class and rely on problem-specific surrogate losses, which hinders learning across classes within a unified framework. In this work, we propose UniHetCO, a unified heterogeneous graph representation for constrained quadratic programming-based combinatorial optimization that encodes problem structure, objective terms, and linear constraints in a single input. This formulation enables training a single model across multiple problem classes with a unified label-free objective. To improve stability under multi-problem learning, we employ a gradient-norm-based dynamic weighting scheme that alleviates gradient imbalance among classes. Experiments on multiple datasets and four constrained problem classes demonstrate competitive performance with state-of-the-art unsupervised NCO baselines, strong cross-problem adaptation potential, and effective warm starts for a commercial classical solver under tight time limits.

cs.LG↗

arXiv2025

Cross-Problem Parameter Transfer in Quantum Approximate Optimization Algorithm: A Machine Learning Approach

Quantum Approximate Optimization Algorithm (QAOA) is one of the most promising candidates to achieve the quantum advantage in solving combinatorial optimization problems. The process of finding a good set of variational parameters in the QAOA circuit has proven to be challenging due to multiple factors, such as barren plateaus. As a result, there is growing interest in exploiting parameter transferability, where parameter sets optimized for one problem instance are transferred to another that could be more complex either to estimate the solution or to serve as a warm start for further optimization. But can we transfer parameters from one class of problems to another? Leveraging parameter sets learned from a well-studied class of problems could help navigate the less studied one, reducing optimization overhead and mitigating performance pitfalls. In this paper, we study whether pretrained QAOA parameters of MaxCut can be used as is or to warm start the Maximum Independent Set (MIS) circuits. Specifically, we design machine learning models to find good donor candidates optimized on MaxCut and apply their parameters to MIS acceptors. Our experimental results show that such parameter transfer can significantly reduce the number of optimization iterations required while achieving comparable approximation ratios.

quant-ph↗

arXiv2025

Interpretable Failure Detection with Human-Level Concepts

Reliable failure detection holds paramount importance in safety-critical applications. Yet, neural networks are known to produce overconfident predictions for misclassified samples. As a result, it remains a problematic matter as existing confidence score functions rely on category-level signals, the logits, to detect failures. This research introduces an innovative strategy, leveraging human-level concepts for a dual purpose: to reliably detect when a model fails and to transparently interpret why. By integrating a nuanced array of signals for each category, our method enables a finer-grained assessment of the model's confidence. We present a simple yet highly effective approach based on the ordinal ranking of concept activation to the input image. Without bells and whistles, our method significantly reduce the false positive rate across diverse real-world image classification benchmarks, specifically by 3.7% on ImageNet and 9% on EuroSAT.

cs.CV↗

arXiv2025

Beyond Accuracy: On the Effects of Fine-tuning Towards Vision-Language Model's Prediction Rationality

Vision-Language Models (VLMs), such as CLIP, have already seen widespread applications. Researchers actively engage in further fine-tuning VLMs in safety-critical domains. In these domains, prediction rationality is crucial: the prediction should be correct and based on valid evidence. Yet, for VLMs, the impact of fine-tuning on prediction rationality is seldomly investigated. To study this problem, we proposed two new metrics called Prediction Trustworthiness and Inference Reliability. We conducted extensive experiments on various settings and observed some interesting phenomena. On the one hand, we found that the well-adopted fine-tuning methods led to more correct predictions based on invalid evidence. This potentially undermines the trustworthiness of correct predictions from fine-tuned VLMs. On the other hand, having identified valid evidence of target objects, fine-tuned VLMs were more likely to make correct predictions. Moreover, the findings are also consistent under distributional shifts and across various experimental settings. We hope our research offer fresh insights to VLM fine-tuning.

cs.LG↗

arXiv2024

SeafloorAI: A Large-scale Vision-Language Dataset for Seafloor Geological Survey

A major obstacle to the advancements of machine learning models in marine science, particularly in sonar imagery analysis, is the scarcity of AI-ready datasets. While there have been efforts to make AI-ready sonar image dataset publicly available, they suffer from limitations in terms of environment setting and scale. To bridge this gap, we introduce SeafloorAI, the first extensive AI-ready datasets for seafloor mapping across 5 geological layers that is curated in collaboration with marine scientists. We further extend the dataset to SeafloorGenAI by incorporating the language component in order to facilitate the development of both vision- and language-capable machine learning models for sonar imagery. The dataset consists of 62 geo-distributed data surveys spanning 17,300 square kilometers, with 696K sonar images, 827K annotated segmentation masks, 696K detailed language descriptions and approximately 7M question-answer pairs. By making our data processing source code publicly available, we aim to engage the marine science community to enrich the data pool and inspire the machine learning community to develop more robust models. This collaborative approach will enhance the capabilities and applications of our datasets within both fields.

cs.CV↗

arXiv2024

Adaptive Cascading Network for Continual Test-Time Adaptation

We study the problem of continual test-time adaption where the goal is to adapt a source pre-trained model to a sequence of unlabelled target domains at test time. Existing methods on test-time training suffer from several limitations: (1) Mismatch between the feature extractor and classifier; (2) Interference between the main and self-supervised tasks; (3) Lack of the ability to quickly adapt to the current distribution. In light of these challenges, we propose a cascading paradigm that simultaneously updates the feature extractor and classifier at test time, mitigating the mismatch between them and enabling long-term model adaptation. The pre-training of our model is structured within a meta-learning framework, thereby minimizing the interference between the main and self-supervised tasks and encouraging fast adaptation in the presence of limited unlabelled data. Additionally, we introduce innovative evaluation metrics, average accuracy and forward transfer, to effectively measure the model's adaptation capabilities in dynamic, real-world scenarios. Extensive experiments and ablation studies demonstrate the superiority of our approach in a range of tasks including image classification, text classification, and speech recognition.

cs.LG↗

arXiv2023

Deep learning-based estimation of whole-body kinematics from multi-view images

It is necessary to analyze the whole-body kinematics (including joint locations and joint angles) to assess risks of fatal and musculoskeletal injuries in occupational tasks. Human pose estimation has gotten more attention in recent years as a method to minimize the errors in determining joint locations. However, the joint angles are not often estimated, nor is the quality of joint angle estimation assessed. In this paper, we presented an end-to-end approach on direct joint angle estimation from multi-view images. Our method leveraged the volumetric pose representation and mapped the rotation representation to a continuous space where each rotation was uniquely represented. We also presented a new kinematic dataset in the domain of residential roofing with a data processing pipeline to generate necessary annotations for the supervised training procedure on direct joint angle estimation. We achieved a mean angle error of $7.19^\circ$ on the new Roofing dataset and $8.41^\circ$ on the Human3.6M dataset, paving the way for employment of on-site kinematic analysis using multi-view images.

cs.CV↗