arXiv subjects

YangQuan Chen

Publications and source records attributed to YangQuan Chen.

At least 19 recordsLinked to original sources

arXiv2026

Information-Guided Safe Reinforcement Learning for Autonomous Gas Source Localization using sUAS

The autonomous localization of fugitive gas emissions using small Unmanned Aircraft Systems (sUAS) constitutes a fundamentally ill-posed inverse problem. In turbulent atmospheric boundary layers, highly intermittent scalar concentration fields violate the assumptions of classical gradient-based navigation, causing data-driven estimators to suffer from severe noise and spurious local minima. To address these challenges, we introduce an Information-Guided Safe Reinforcement Learning framework evaluated within a custom, GPU-accelerated 3D simulation environment coupling an Eulerian wind solver with a Lagrangian puff dispersion model. We identify a critical vulnerability in deterministic information-seeking planners - a Gramian bias where agents act greedily upon flawed early estimates, starving the estimator of spatial diversity. To systematically break this degeneracy, our architecture integrates a classical empirical observability Gramian (EMGR) planner with a learned Soft Actor-Critic (SAC) exploratory policy. A deterministic meta-supervisor actively monitors estimator reliability via Kullback-Leibler (KL) divergence, dynamically blending deterministic exploitation with learned exploration to steer the sUAS into high-information zones. Trained via a progressive curriculum and safeguarded by a strictly enforced Robust Control Barrier Function (RCBF), our RL framework achieves nearly 80% localization success on complex, mobile sources - drastically outperforming classical baselines (~30%) - while ensuring zero safety violations.

cs.RO↗

arXiv2026

Fractional-Order Federated Learning

Federated learning (FL) allows remote clients to train a global model collaboratively while protecting client privacy. Despite its privacy-preserving benefits, FL has significant drawbacks, including slow convergence, high communication cost, and non-independent-and-identically-distributed (non-IID) data. In this work, we present a novel FedAvg variation called Fractional-Order Federated Averaging (FOFedAvg), which incorporates Fractional-Order Stochastic Gradient Descent (FOSGD) to capture long-range relationships and deeper historical information. By introducing memory-aware fractional-order updates, FOFedAvg improves communication efficiency and accelerates convergence while mitigating instability caused by heterogeneous, non-IID client data. We compare FOFedAvg against a broad set of established federated optimization algorithms on benchmark datasets including MNIST, FEMNIST, CIFAR-10, CIFAR-100, EMNIST, the Cleveland heart disease dataset, Sent140, PneumoniaMNIST, and Edge-IIoTset. Across a range of non-IID partitioning schemes, FOFedAvg is competitive with, and often outperforms, these baselines in terms of test performance and convergence speed. On the theoretical side, we prove that FOFedAvg converges to a stationary point under standard smoothness and bounded-variance assumptions for fractional order $0<\alpha\le 1$. Together, these results show that fractional-order, memory-aware updates can substantially improve the robustness and effectiveness of federated learning, offering a practical path toward distributed training on heterogeneous data.

cs.LG↗

arXiv2026

Fractional Order Federated Learning for Battery Electric Vehicle Energy Consumption Modeling

Federated learning on connected electric vehicles (BEVs) faces severe instability due to intermittent connectivity, time-varying client participation, and pronounced client-to-client variation induced by diverse operating conditions. Conventional FedAvg and many advanced methods can suffer from excessive drift and degraded convergence under these realistic constraints. This work introduces Fractional-Order Roughness-Informed Federated Averaging (FO-RI-FedAvg), a lightweight and modular extension of FedAvg that improves stability through two complementary client-side mechanisms: (i) adaptive roughness-informed proximal regularization, which dynamically tunes the pull toward the global model based on local loss-landscape roughness, and (ii) non-integer-order local optimization, which incorporates short-term memory to smooth conflicting update directions. The approach preserves standard FedAvg server aggregation, adds only element-wise operations with amortizable overhead, and allows independent toggling of each component. Experiments on two real-world BEV energy prediction datasets, VED and its extended version eVED, show that FO-RI-FedAvg achieves improved accuracy and more stable convergence compared to strong federated baselines, particularly under reduced client participation.

cs.LG↗

arXiv2026

When Gradient Clipping Becomes a Control Mechanism for Differential Privacy in Deep Learning

Privacy-preserving training on sensitive data commonly relies on differentially private stochastic optimization with gradient clipping and Gaussian noise. The clipping threshold is a critical control knob: if set too small, systematic over-clipping induces optimization bias; if too large, injected noise dominates updates and degrades accuracy. Existing adaptive clipping methods often depend on per-example gradient norm statistics, adding computational overhead and introducing sensitivity to datasets and architectures. We propose a control-driven clipping strategy that adapts the threshold using a lightweight, weight-only spectral diagnostic computed from model parameters. At periodic probe steps, the method analyzes a designated weight matrix via spectral decomposition and estimates a heavy-tailed spectral indicator associated with training stability. This indicator is smoothed over time and fed into a bounded feedback controller that updates the clipping threshold multiplicatively in the log domain. Because the controller uses only parameters produced during privacy-preserving training, the resulting threshold updates are post-processing and do not increase privacy loss beyond that of the underlying DP optimizer under standard composition accounting.

cs.LG↗

arXiv2026

Roughness-Informed Federated Learning

Federated Learning (FL) enables collaborative model training across distributed clients while preserving data privacy, yet faces challenges in non-independent and identically distributed (non-IID) settings due to client drift, which impairs convergence. We propose RI-FedAvg, a novel FL algorithm that mitigates client drift by incorporating a Roughness Index (RI)-based regularization term into the local objective, adaptively penalizing updates based on the fluctuations of local loss landscapes. This paper introduces RI-FedAvg, leveraging the RI to quantify the roughness of high-dimensional loss functions, ensuring robust optimization in heterogeneous settings. We provide a rigorous convergence analysis for non-convex objectives, establishing that RI-FedAvg converges to a stationary point under standard assumptions. Extensive experiments on MNIST, CIFAR-10, and CIFAR-100 demonstrate that RI-FedAvg outperforms state-of-the-art baselines, including FedAvg, FedProx, FedDyn, SCAFFOLD, and DP-FedAvg, achieving higher accuracy and faster convergence in non-IID scenarios. Our results highlight RI-FedAvg's potential to enhance the robustness and efficiency of federated learning in practical, heterogeneous environments.

cs.LG↗

arXiv2026

Statistical Roughness-Informed Machine Unlearning

Machine unlearning aims to remove the influence of a designated forget set from a trained model while preserving utility on the retained data. In modern deep networks, approximate unlearning frequently fails under large or adversarial deletions due to pronounced layer-wise heterogeneity: some layers exhibit stable, well-regularized representations while others are brittle, undertrained, or overfit, so naive update allocation can trigger catastrophic forgetting or unstable dynamics. We propose Statistical-Roughness Adaptive Gradient Unlearning (SRAGU), a mechanism-first unlearning algorithm that reallocates unlearning updates using layer-wise statistical roughness operationalized via heavy-tailed spectral diagnostics of layer weight matrices. Starting from an Adaptive Gradient Unlearning (AGU) sensitivity signal computed on the forget set, SRAGU estimates a WeightWatcher-style heavy-tailed exponent for each layer, maps it to a bounded spectral stability weight, and uses this stability signal to spectrally reweight the AGU sensitivities before applying the same minibatch update form. This concentrates unlearning motion in spectrally stable layers while damping updates in unstable or overfit layers, improving stability under hard deletions. We evaluate unlearning via behavioral alignment to a gold retrained reference model trained from scratch on the retained data, using empirical prediction-divergence and KL-to-gold proxies on a forget-focused query set; we additionally report membership inference auditing as a complementary leakage signal, treating forget-set points as should-be-forgotten members during evaluation.

cs.LG↗

arXiv2025

More Optimal Fractional-Order Stochastic Gradient Descent for Non-Convex Optimization Problems

Fractional-order stochastic gradient descent (FOSGD) leverages fractional exponents to capture long-memory effects in optimization. However, its utility is often limited by the difficulty of tuning and stabilizing these exponents. We propose 2SED Fractional-Order Stochastic Gradient Descent (2SEDFOSGD), which integrates the Two-Scale Effective Dimension (2SED) algorithm with FOSGD to adapt the fractional exponent in a data-driven manner. By tracking model sensitivity and effective dimensionality, 2SEDFOSGD dynamically modulates the exponent to mitigate oscillations and hasten convergence. Theoretically, for onoconvex optimization problems, this approach preserves the advantages of fractional memory without the sluggish or unstable behavior observed in na\"ive fractional SGD. Empirical evaluations in Gaussian and $\alpha$-stable noise scenarios using an autoregressive (AR) model highlight faster convergence and more robust parameter estimates compared to baseline methods, underscoring the potential of dimension-aware fractional techniques for advanced modeling and estimation tasks.

cs.LG↗

arXiv2025

Controlled Optimization with a Prescribed Finite-Time Convergence Using a Time Varying Feedback Gradient Flow

From the perspective of control theory, the gradient descent optimization methods can be regarded as a dynamic system where various control techniques can be designed to enhance the performance of the optimization method. In this paper, we propose a prescribed finite-time convergent gradient flow that uses time-varying gain nonlinear feedback that can drive the states smoothly towards the minimum. This idea is different from the traditional finite-time convergence algorithms that relies on fractional-power or signed gradient as a nonlinear feedback, that is proved to have finite/fixed time convergence satisfying strongly convex or the Polyak-{\L}ojasiewicz (P{\L}) inequality, where due to its nature, the proposed approach was shown to achieve this property for both strongly convex function, and for those satisfies Polyak-{\L}ojasiewic inequality. Our method is proved to converge in a prescribed finite time via Lyapunov theory. Numerical experiments were presented to illustrate our results.

math.OC↗

arXiv2025

Effective Dimension Aware Fractional-Order Stochastic Gradient Descent for Convex Optimization Problems

Fractional-order stochastic gradient descent (FOSGD) leverages fractional exponents to capture long-memory effects in optimization. However, its utility is often limited by the difficulty of tuning and stabilizing these exponents. We propose 2SED Fractional-Order Stochastic Gradient Descent (2SEDFOSGD), which integrates the Two-Scale Effective Dimension (2SED) algorithm with FOSGD to adapt the fractional exponent in a data-driven manner. By tracking model sensitivity and effective dimensionality, 2SEDFOSGD dynamically modulates the exponent to mitigate oscillations and hasten convergence. Theoretically, this approach preserves the advantages of fractional memory without the sluggish or unstable behavior observed in na\"ive fractional SGD. Empirical evaluations in Gaussian and $\alpha$-stable noise scenarios using an autoregressive (AR) model\textcolor{red}{, as well as on the MNIST and CIFAR-100 datasets for image classification,} highlight faster convergence and more robust parameter estimates compared to baseline methods, underscoring the potential of dimension-aware fractional techniques for advanced modeling and estimation tasks.

cs.LG↗

arXiv2021

Finite-time and Fixed-time Convergence in Continuous-time Optimization

It is known that the gradient method can be viewed as a dynamic system where various iterative schemes can be designed as a part of the closed loop system with desirable properties. In this paper, the finite-time and fixed-time convergence in continuous-time optimization are mainly considered. By the advantage of sliding mode control, a finite-time gradient method is proposed, whose convergence time is dependent on initial conditions. To make the convergence time robust to initial conditions, two different designs of fixed-time gradient methods are then provided. One is designed using the property of sine function, whose convergence time is dependent on the frequency of a sine function. The other one is designed using the property of Mittag-Leffler function, whose convergence time is determined by the first positive zero of a Mittag-Leffler function. All the results are extended to more general cases and finally demonstrated by some dedicated simulation examples.

math.OC↗

arXiv2021

Optimal Regional Tracking Control of Time-Fractional Diffusion Systems

In this paper, we aim to explore optimal regional trajectory tracking control problems of the anomalous subdiffusion processes governed by time-fractional diffusion systems under the Neumann boundary conditions. Using eigenvalue theory of the system operator and the semigroup theory, we explore the existence and some estimates of the mild solution to the considered system. An approach on finding solution to the optimal problem that minimizes the regional trajectory tracking error and the corresponding control cost over a finite space and time domain is then explored via the Hilbert uniqueness method (HUM). The obtained results not only can be directly used to investigate the systems that are not controllable on the whole domain, but also yield an explicit expression of the control signal in terms of the desired trajectory. Most importantly, it is worth noting that our results in this paper are still novel even for the special case when the order of fractional derivative is equal to one. Finally, we provide a numerical example to illustrate our theoretical results.

math.OC↗

arXiv2021

Self-optimizing loop sifting and majorization for 3D reconstruction

Visual simultaneous localization and mapping (vSLAM) and 3D reconstruction methods have gone through impressive progress. These methods are very promising for autonomous vehicle and consumer robot applications because they can map large-scale environments such as cities and indoor environments without the need for much human effort. However, when it comes to loop detection and optimization, there is still room for improvement. vSLAM systems tend to add the loops very conservatively to reduce the severe influence of the false loops. These conservative checks usually lead to correct loops rejected, thus decrease performance. In this paper, an algorithm that can sift and majorize loop detections is proposed. Our proposed algorithm can compare the usefulness and effectiveness of different loops with the dense map posterior (DMP) metric. The algorithm tests and decides the acceptance of each loop without a single user-defined threshold. Thus it is adaptive to different data conditions. The proposed method is general and agnostic to sensor type (as long as depth or LiDAR reading presents), loop detection, and optimization methods. Neither does it require a specific type of SLAM system. Thus it has great potential to be applied to various application scenarios. Experiments are conducted on public datasets. Results show that the proposed method outperforms state-of-the-art methods.

cs.CV↗

arXiv2021

Consistent approximation of fractional order operators

Fractional order controllers become increasingly popular due to their versatility and superiority in various performance. However, the bottleneck in deploying these tools in practice is related to their analog or numerical implementation. Numerical approximations are usually employed in which the approximation of fractional differintegrator is the foundation. Generally, the following three identical equations always hold, i.e., $\frac{1}{s^\alpha}\frac{1}{s^{1-\alpha}} = \frac{1}{s}$, $s^\alpha \frac{1}{s^\alpha} = 1$ and $s^\alpha s^{1-\alpha} = s$. However, for the approximate models of fractional differintegrator $s^\alpha$, $\alpha\in(-1,0)\cup(0,1)$, there usually exist some conflicts on the mentioned equations, which might enlarge the approximation error or even cause fallacies in multiple orders occasion. To overcome the conflicts, this brief develops a piecewise approximate model and provides two procedures for designing the model parameters. The comparison with several existing methods shows that the proposed methods do not only satisfy the equalities but also achieve high approximation accuracy. From this, it is believed that this work can serve for simulation and realization of fractional order controllers more friendly.

math.NA↗

arXiv2021

A metric for evaluating 3D reconstruction and mapping performance with no ground truthing

It is not easy when evaluating 3D mapping performance because existing metrics require ground truth data that can only be collected with special instruments. In this paper, we propose a metric, dense map posterior (DMP), for this evaluation. It can work without any ground truth data. Instead, it calculates a comparable value, reflecting a map posterior probability, from dense point cloud observations. In our experiments, the proposed DMP is benchmarked against ground truth-based metrics. Results show that DMP can provide a similar evaluation capability. The proposed metric makes evaluating different methods more flexible and opens many new possibilities, such as self-supervised methods and more available datasets.

cs.CV↗

arXiv2020

More Informed Random Sample Consensus

Random sample consensus (RANSAC) is a robust model-fitting algorithm. It is widely used in many fields including image-stitching and point cloud registration. In RANSAC, data is uniformly sampled for hypothesis generation. However, this uniform sampling strategy does not fully utilize all the information on many problems. In this paper, we propose a method that samples data with a L\'{e}vy distribution together with a data sorting algorithm. In the hypothesis sampling step of the proposed method, data is sorted with a sorting algorithm we proposed, which sorts data based on the likelihood of a data point being in the inlier set. Then, hypotheses are sampled from the sorted data with L\'{e}vy distribution. The proposed method is evaluated on both simulation and real-world public datasets. Our method shows better results compared with the uniform baseline method.

cs.RO↗

arXiv2020

FaultFace: Deep Convolutional Generative Adversarial Network (DCGAN) based Ball-Bearing Failure Detection Method

Failure detection is employed in the industry to improve system performance and reduce costs due to unexpected malfunction events. So, a good dataset of the system is desirable for designing an automated failure detection system. However, industrial process datasets are unbalanced and contain little information about failure behavior due to the uniqueness of these events and the high cost for running the system just to get information about the undesired behaviors. For this reason, performing correct training and validation of automated failure detection methods is challenging. This paper proposes a methodology called FaultFace for failure detection on Ball-Bearing joints for rotational shafts using deep learning techniques to create balanced datasets. The FaultFace methodology uses 2D representations of vibration signals denominated faceportraits obtained by time-frequency transformation techniques. From the obtained faceportraits, a Deep Convolutional Generative Adversarial Network is employed to produce new faceportraits of the nominal and failure behaviors to get a balanced dataset. A Convolutional Neural Network is trained for fault detection employing the balanced dataset. The FaultFace methodology is compared with other deep learning techniques to evaluate its performance in for fault detection with unbalanced datasets. Obtained results show that FaultFace methodology has a good performance for failure detection for unbalanced datasets.

eess.IV↗

arXiv2020

PHELP: Pixel Heating Experiment Learning Platform for Education and Research on IAI-based Smart Control Engineering

Thermal processes are one of the most common systems in the industry, making its understanding a mandatory skill for control engineers. So, multiple efforts are focused on developing low-cost and portable experimental training rigs recreating the thermal process dynamics and controls, usually limited to SISO or low order 2x2 MIMO systems. This paper presents PHELP, a low-cost, portable, and high order MIMO educational platform for uniformity temperature control training. The platform is composed of an array of 16 Peltier modules as heating elements, with a lower heating and cooling times, resulting in a 16x16 high order MIMO system. A low-cost real-time infrared thermal camera is employed as a temperature feedback sensor instead of a standard thermal sensor, ideal for high order MIMO system sensing and temperature distribution tracking. The control algorithm is developed in Matlab/Simulink and employs an Arduino board in hardware in the loop configuration to apply the control action to each Peltier module in the array. A temperature control experiment is performed, showing that the platform is suitable for teaching and training experiences not only in the classroom but also for engineers in the industry. Furthermore, various abnormal conditions can be introduced so that smart control engineering features can be tested.

eess.SY↗

arXiv2020

Information-Based Model Discrimination for Digital Twin Behavioral Matching

Digital Twin is a breaking technology that allows creating virtual representations of complex physical systems based on updated information of the system and its physical laws. However, making the Digital Twin behavior matching with the real system can be challenging due to the number of unknown parameters in each twin. Its search can be done using optimization-based techniques, producing a family of models based on different system datasets, so, a discrimination criterion is required to determine the best Digital Twin model. This paper presents an information theory-based discrimination criterion to determine the best Digital Twin model resulting from a behavioral matching process. The information gain of a model is employed as a discrimination criterion. Box-Jenkins models are used to define the family of models for each behavioral matching result. The proposed method is compared with other information-based metrics as well as the $\nu$gap metric. As a study case, the discrimination method is applied to the Digital Twin for a real-time vision feedback infrared temperature uniformity control system. Obtained results show that information-based methodologies are useful for selecting an accurate Digital Twin model representing the system among a family of plants

eess.SP↗