Search Results (43)

Personalized Federated Learning (PFL) seeks to train client-specific models across distributed data silos with heterogeneous distributions. We introduce Sign-Entropy Regularization (SER), a novel entropy-based regularization technique that penalizes excessive directional variability in client-local optimization. Motivated by Descartes’ Rule of Signs, we hypothesize that frequent sign changes in gradient trajectories reflect complexity in the local loss landscape. By minimizing the entropy of gradient sign patterns during local updates, SER encourages smoother optimization paths, improves convergence stability, and enhances personalization. We formally define a differentiable sign-entropy objective over the gradient sign distribution and integrate it into standard federated optimization frameworks, including FedAvg and FedProx. The regularizer is computed efficiently and applied post hoc per local round. Extensive experiments on three benchmark datasets (FEMNIST, Shakespeare, and CIFAR-10) show that SER improves both average and worst-case client accuracy, reduces variance across clients, accelerates convergence, and smooths the local loss surface as measured by Hessian trace and spectral norm. We also present a sensitivity analysis of the regularization strength $\rho$ and discuss the potential for client-adaptive variants. Comparative evaluations against state-of-the-art methods (e.g., Ditto, pFedMe, momentum-based variants, Entropy-SGD) highlight that SER introduces an orthogonal and scalable mechanism for personalization. Theoretically, we frame SER as an information-theoretic and geometric regularizer that stabilizes learning dynamics without requiring dual-model structures or communication modifications. This work opens avenues for trajectory-based regularization and hybrid entropy-guided optimization in federated and resource-constrained learning settings. Full article

This study presents a novel artificial intelligence-driven architecture for real-time sensor fusion in autonomous vehicles (AVs), leveraging Apache Kafka and MongoDB for synchronous and asynchronous data processing to enhance resilience against sensor failures and dynamic conditions. We introduce Chaotic AdaBoost (CAB), an advanced variant of AdaBoost that integrates a logistic chaotic map into its weight update process, overcoming the limitations of deterministic ensemble methods. CAB is evaluated alongside k-Nearest Neighbors (kNNs), Artificial Neural Networks (ANNs), standard AdaBoost (AB), Gradient Boosting (GBa), and Random Forest (RF) for speed and acceleration prediction using CARLA simulator data. CAB achieves a superior 99.3% accuracy (MSE: 0.018 for acceleration, 0.010 for speed; MAE: 0.020 for acceleration, 0.012 for speed; R²: 0.993 for acceleration, 0.997 for speed), a mean Time-To-Collision (TTC) of 3.2 s, and jerk of 0.15 m/s³, outperforming AB (98.5%, MSE: 0.15, TTC: 2.8 s, jerk: 0.22 m/s³), GB (99.1%), ANN (98.2%), RF (97.5%), and kNN (87.0%). This logistic map-enhanced adaptability, reducing MSE by 88% over AB, ensures robust anomaly detection and data fusion under uncertainty, critical for AV safety and comfort. Despite a 20% increase in training time (72 s vs. 60 s for AB), CAB’s integration with Kafka’s high-throughput streaming maintains real-time efficacy, offering a scalable framework that advances operational reliability and passenger experience in autonomous driving. Full article

This study proposes a robust methodology for vibration suppression and trajectory tracking in rotary flexible-link systems by leveraging guided reinforcement learning (GRL). The approach integrates the twin delayed deep deterministic policy gradient (TD3) algorithm with a linear quadratic regulator (LQR) acting as a guiding controller during training. Flexible-link mechanisms common in advanced robotics and aerospace systems exhibit oscillatory behavior that complicates precise control. To address this, the system is first identified using experimental input-output data from a Quanser^® virtual plant, generating an accurate state-space representation suitable for simulation-based policy learning. The hybrid control strategy enhances sample efficiency and accelerates convergence by incorporating LQR-generated trajectories during TD3 training. Internally, the TD3 agent benefits from architectural features such as twin critics, delayed policy updates, and target action smoothing, which collectively improve learning stability and reduce overestimation bias. Comparative results show that the guided TD3 controller achieves superior performance in terms of vibration damping, transient response, and robustness, when compared to conventional LQR, fuzzy logic, neural networks, and GA-LQR approaches. Although the controller was validated using a high-fidelity digital twin, it has not yet been deployed on the physical plant. Future work will focus on real-time implementation and structural robustness testing under parameter uncertainty. Overall, this research demonstrates that guided reinforcement learning can yield stable and interpretable policies that comply with classical control criteria, offering a scalable and generalizable framework for intelligent control of flexible mechanical systems. Full article

This study presents the first investigation into the problem of autonomous vehicle (AV) merging into existing platoons, proposing a multi-agent deep reinforcement learning (MA-DRL)-based cooperative control framework. The developed MA-DRL architecture enables coordinated learning among multiple autonomous agents to address the multi-objective coordination challenge through synchronized control of platoon longitudinal acceleration, AV steering and acceleration. To enhance training efficiency, we develop a dual-layer multi-agent maximum Q-value proximal policy optimization (MAMQPPO) method, which extends the multi-agent PPO algorithm (a policy gradient method ensuring stable policy updates) by incorporating maximum Q-value action selection for platoon gap control and discrete command generation. This method simplifies the training process by using maximum Q-value action policy optimization to learn platoon gap selection and discrete action commands. Furthermore, a partially decoupled reward function (PD-Reward) is designed to properly guide the behavioral actions of both AVs and platoons while accelerating network convergence. Comprehensive highway simulation experiments show the proposed method reduces merging time by 37.69% (12.4 s vs. 19.9 s) and energy consumption by 58% (3.56 kWh vs. 8.47 kWh) compared to existing methods (the quintic polynomial-based + PID (Proportional–Integral–Differential)). Full article

Aiming at the problem of nonlinear coupling error in the measurement of parallel six-axis accelerometers, this study improves the back propagation (BP) neural network and proposes an improved BP neural network decoupling model that introduces the gradient descent with momentum and the Levenberg–Marquardt (LM) algorithm. By introducing the momentum factor in the model updating stage, the LM algorithm is used in the local learning stage to improve the convergence speed and shock resistance of the network, and to enhance the accuracy of the algorithm. Based on the mid-frequency standard vibration device APS 129 ELECTRO-SEIS (SPEKTRA, Stuttgart, Baden-Württemberg, Germany), the calibration data are obtained and the improved BP neural network decoupling model is trained to complete the nonlinear decoupling of the test set. Compared with the linear decoupling method, the decoupled six-axis accelerometers with the improved BP neural network model have acceleration measurement accuracies of 0.035%, 0.018% and 0.039% along the x, y and z axes, respectively, which indicates that the model has high decoupling accuracy, and it can significantly improve the measurement accuracy of the sensors. The research results can provide theoretical support for high-precision inertial navigation. Full article

This paper proposes an improved parameter extraction optimization algorithm for radio frequency (RF) devices. The algorithm integrates parameter classification and correction, gradient-based performance handling, bias-aware updates, and group-based optimization strategies, achieving enhanced optimization accuracy, accelerated convergence, and improved stability. It effectively addresses the limitations of deterministic algorithms in RF device parameter extraction optimization, such as low efficiency, sensitivity to initial values, and unstable convergence. To validate the algorithm’s effectiveness, a Ka-band filter performance curve fitting case study was conducted. By comparing simulated curves with optimized fitted curves, the advantages of the algorithm in terms of optimization efficiency, accuracy, and convergence stability were demonstrated. Experimental results show that, compared to traditional optimization algorithms, the proposed method significantly improves curve fitting accuracy, computational efficiency, and stability, highlighting its application value in RF device parameter extraction. Full article

In this paper, based on the novel generalized Hamilton-real (GHR) calculus, we propose for the first time a quaternion Nesterov accelerated projected gradient algorithm for computing the dominant eigenvalue and eigenvector of quaternion Hermitian matrices. By introducing momentum terms and look-ahead updates, the algorithm achieves a faster convergence rate. We theoretically prove the convergence of the quaternion Nesterov accelerated projected gradient algorithm. Numerical experiments show that the proposed method outperforms the quaternion projected gradient ascent method and the traditional algebraic methods in terms of computational accuracy and runtime efficiency. Full article

In the contemporary landscape of healthcare, machine learning models are pivotal in facilitating precise predictions, particularly in the nuanced diagnosis of complex ailments such as breast cancer. Traditional diagnostic methodologies grapple with inherent challenges, including excessive complexity, elevated costs, and reliance on subjective interpretation, which frequently culminate in inaccuracies. The urgency of early detection cannot be overstated, as it markedly broadens treatment modalities and significantly enhances survival rates. This paper delineates an innovative optimization framework designed to augment diagnostic accuracy by amalgamating momentum-based optimization techniques within a neural network paradigm. Conventional machine learning approaches are often encumbered by issues of overfitting, data imbalance, and the inadequacy of capturing intricate patterns in high-dimensional datasets. To counter these limitations, we propose a sophisticated framework that integrates an adaptive threshold mechanism across an array of gradient-based optimizers, including SGD, RMSprop, adam, adagrad, adamax, adadelta, nadam and Nesterov momentum. This novel approach effectively mitigates oscillatory behavior, refines parameter updates, and accelerates convergence. A salient feature of our methodology is the incorporation of a momentum threshold for early stopping, which ceases training upon the stabilization of momentum below a pre-defined threshold, thereby pre-emptively preventing overfitting. Leveraging the Wisconsin Breast Cancer Dataset, our model achieved a remarkable 99.72% accuracy and 100% sensitivity, significantly curtailing misclassification rates compared to traditional methodologies. This framework stands as a robust solution for early breast cancer diagnosis, thereby enhancing clinical decision making and improving patient outcomes. Full article

Recently, developing multi-UAVs to cooperatively pursue a fast-moving target has become a research hotspot in the current world. Although deep reinforcement learning (DRL) has made a lot of achievements in the UAV pursuit game, there are still some problems such as high-dimensional parameter space, the ease of falling into local optimization, the long training time, and the low task success rate. To solve the above-mentioned issues, we propose an improved twin delayed deep deterministic policy gradient algorithm combining the genetic algorithm and maximum mean discrepancy method (GM-TD3) for multi-UAV cooperative pursuit of high-speed targets. Firstly, this paper combines GA-based evolutionary strategies with TD3 to generate action networks. Then, in order to avoid local optimization in the algorithm training process, the maximum mean difference (MMD) method is used to increase the diversity of the policy population in the updating process of the population parameters. Finally, by setting the sensitivity weights of the genetic memory buffer of UAV individuals, the mutation operator is improved to enhance the stability of the algorithm. In addition, this paper designs a hybrid reward function to accelerate the convergence speed of training. Through simulation experiments, we have verified that the training efficiency of the improved algorithm has been greatly improved, which can achieve faster convergence; the successful rate of the task has reached 95%, and further validated UAVs can better cooperate to complete the pursuit game task. Full article

The prediction of the ablation rate of silicone rubber-based composites is of great significance to accelerate the development of flexible thermal protection materials. Herein, a method which combines uniform design experimentation, active learning, and virtual sample generation was proposed to establish a prediction model of the mass ablation rate based on a small dataset. Briefly, a small number of sample points were collected using uniform design experimentation, which were marked to construct the initial dataset and primitive model. Then, data points were acquired from the sample pool and iterated using various integrated algorithms through active learning to update the above dataset and model. Finally, a large number of virtual samples were generated based on the optimal model, and a further optimized prediction model was achieved. The results showed that after introducing 300 virtual samples, the average percentage error of the gradient boosting decision tree (GBDT) prediction model on the test set decreased to 3.1%, which demonstrates the effectiveness of the proposed method in building prediction models based on a small dataset. Full article

Particle-based Variational Inference (ParVI) methods have been widely adopted in deep Bayesian inference tasks such as Bayesian neural networks or Gaussian Processes, owing to their efficiency in generating high-quality samples given the score of the target distribution. Typically, ParVI methods evolve a weighted-particle system by approximating the first-order Wasserstein gradient flow to reduce the dissimilarity between the particle system’s empirical distribution and the target distribution. Recent advancements in ParVI have explored sophisticated gradient flows to obtain refined particle systems with either accelerated position updates or dynamic weight adjustments. In this paper, we introduce the semi-Hamiltonian gradient flow on a novel Information–Fisher–Rao space, known as the SHIFR flow, and propose the first ParVI framework that possesses both accelerated position update and dynamical weight adjustment simultaneously, named the General Accelerated Dynamic-Weight Particle-based Variational Inference (GAD-PVI) framework. GAD-PVI is compatible with different dissimilarities between the empirical distribution and the target distribution, as well as different approximation approaches to gradient flow. Moreover, when the appropriate dissimilarity is selected, GAD-PVI is also suitable for obtaining high-quality samples even when analytical scores cannot be obtained. Experiments conducted under both the score-based tasks and sample-based tasks demonstrate the faster convergence and reduced approximation error of GAD-PVI methods over the state-of-the-art. Full article

This paper applies federated reinforcement learning (FRL) in cellular vehicle-to-everything (C-V2X) communication to enable vehicles to learn communication parameters in collaboration with a parameter server that is embedded in an unmanned aerial vehicle (UAV). Different sensors in vehicles capture different types of data, contributing to data heterogeneity. C-V2X communication networks impose additional communication overhead in order to converge to a global model when the sensor data are not independent-and-identically-distributed (non-i.i.d.). Consequently, the training time for local model updates also varies considerably. Using FRL, we accelerated this convergence by minimizing communication rounds, and we delayed it by exploring the correlation between the data captured by various vehicles in subsequent time steps. Additionally, as UAVs have limited battery power, processing of the collected information locally at the vehicles and then transmitting the model hyper-parameters to the UAVs can optimize the available power consumption pattern. The proposed FRL algorithm updates the global model through adaptive weighing of Q-values at each training round. By measuring the local gradients at the vehicle and the global gradient at the UAV, the contribution of the local models is determined. We quantify these Q-values using nonlinear mappings to reinforce positive rewards such that the contribution of local models is dynamically measured. Moreover, minimizing the number of communication rounds between the UAVs and vehicles is investigated as a viable approach for minimizing delay. A performance evaluation revealed that the FRL approach can yield up to a 40% reduction in the number of communication rounds between vehicles and UAVs when compared to gross data offloading. Full article

Electropolishing (EP) has become a standard procedure for treating the inner surfaces of superconducting radio-frequency (SRF) cavities composed of pure niobium. In this study, a new EP facility was employed for the surface treatment of both 1.3 GHz and 3.9 GHz single-cell cavities at the Wuxi Platform. The stable “cold EP” mode was successfully implemented on this newly designed EP facility. By integrating the cold EP process with a two-step baking approach, a maximum accelerating gradient exceeding 40 MV/m was achieved in 1.3 GHz single-cell cavities. Additionally, an update to this EP facility involved the design of a special cathode system for small-aperture structures, facilitating the cold EP process for 3.9 GHz single-cell cavities. Ultimately, a maximum accelerating gradient exceeding 25 MV/m was attained in the 3.9 GHz single-cell cavities after undergoing the cold EP treatment. The design and commissioning of the EP device, as well as the electropolishing and vertical test results of the single-cell cavities, will be detailed herein. These methods and experiences are also transferable to multi-cell cavities and elliptical cavities of other frequencies. Full article

Adversarial example generation is a technique that involves perturbing inputs with imperceptible noise to induce misclassifications in neural networks, serving as a means to assess the robustness of such models. Among the adversarial attack algorithms, momentum iterative fast gradient sign Method (MI-FGSM) and its variants constitute a class of highly effective offensive strategies, achieving near-perfect attack success rates in white-box settings. However, these methods’ use of sign activation functions severely compromises gradient information, which leads to low success rates in black-box attacks and results in large adversarial perturbations. In this paper, we introduce a novel adversarial attack algorithm, NA-FGTM. Our method employs the Tanh activation function instead of the sign which can accurately preserve gradient information. In addition, it utilizes the Adam optimization algorithm as well as the Nesterov acceleration, which is able to stabilize gradient update directions and expedite gradient convergence. Above all, the transferability of adversarial examples can be enhanced. Through integration with data augmentation techniques such as DIM, TIM, and SIM, NA-FGTM can further improve the efficacy of black-box attacks. Extensive experiments on the ImageNet dataset demonstrate that our method outperforms the state-of-the-art approaches in terms of black-box attack success rate and generates adversarial examples with smaller perturbations. Full article

Effective real-time autonomous navigation for mobile robots in static and dynamic environments has become a challenging and active research topic. Although the simultaneous localization and mapping (SLAM) algorithm offers a solution, it often heavily relies on complex global and local maps, resulting in significant computational demands, slower convergence rates, and prolonged training times. In response to these challenges, this paper presents a novel algorithm called PER-n₂D₃QN, which integrates prioritized experience replay, a noisy network with factorized Gaussian noise, n-step learning, and a dueling structure into a double deep Q-network. This combination enhances the efficiency of experience replay, facilitates exploration, and provides more accurate Q-value estimates, thereby significantly improving the performance of autonomous navigation for mobile robots. To further bolster the stability and robustness, meaningful improvements, such as target “soft” updates and the gradient clipping mechanism, are employed. Additionally, a novel and powerful target-oriented reshaping reward function is designed to expedite learning. The proposed model is validated through extensive experiments using the robot operating system (ROS) and Gazebo simulation environment. Furthermore, to more specifically reflect the complexity of the simulation environment, this paper presents a quantitative analysis of the simulation environment. The experimental results demonstrate that PER-n₂D₃QN exhibits heightened accuracy, accelerated convergence rates, and enhanced robustness in both static and dynamic scenarios. Full article