Kernel methods are powerful tools for nonlinear learning with well-established theory. The scalability issue has been their long-standing challenge. Despite the existing success, there are two limitations in large-scale kernel methods: (i) The memory overhead is too high for users to afford; (ii) existing efforts mainly focus on kernel ridge regression (KRR), while other models lack study. In this paper, we propose Joker, a joint optimization framework for diverse kernel models, including KRR, logistic regression, and support vector machines. We design a dual block coordinate descent method with trust region (DBCD-TR) and adopt kernel approximation with randomized features, leading to low memory costs and high efficiency in large-scale learning. Experiments show that Joker saves up to 90% memory but achieves comparable training time and performance (or even better) than the state-of-the-art methods.

Subspace learning and Support Vector Machine (SVM) are two critical techniques in pattern recognition, playing pivotal roles in feature extraction and classification. However, how to learn the optimal subspace such that the SVM classifier can perform the best is still a challenging problem due to the difficulty in optimization, computation, and algorithm convergence. To address these problems, this paper develops a novel method named Optimal Discriminant Support Vector Machine (ODSVM), which integrates support vector classification with discriminative subspace learning in a seamless framework. As a result, the most discriminative subspace and the corresponding optimal SVM are obtained simultaneously to pursue the best classification performance. The efficient optimization framework is designed for binary and multi-class ODSVM. Moreover, a fast sequential minimization optimization (SMO) algorithm with pruning is proposed to accelerate the computation in multi-class ODSVM. Unlike other related methods, ODSVM has a strong theoretical guarantee of global convergence, highlighting its superiority and stability. Numerical experiments are conducted on thirteen datasets and the results demonstrate that ODSVM outperforms existing methods with statistical significance.

Vehicular Metaverses represent emerging paradigms arising from the convergence of vehicle road cooperation, Metaverse, and augmented intelligence of things. Users engaging with Vehicular Metaverses (VMUs) gain entry by consistently updating their Vehicular Twins (VTs), which are deployed on RoadSide Units (RSUs) in proximity. The constrained RSU coverage and the consistently moving vehicles necessitate the continuous migration of VTs between RSUs through vehicle road cooperation, ensuring uninterrupted immersion services for VMUs. Nevertheless, the VT migration process faces challenges in obtaining adequate bandwidth resources from RSUs for timely migration, posing a resource trading problem among RSUs. In this paper, we tackle this challenge by formulating a game-theoretic incentive mechanism with multi-leader multi-follower, incorporating insights from social-awareness and queueing theory to optimize VT migration. To validate the existence and uniqueness of the Stackelberg Equilibrium, we apply the backward induction method. Theoretical solutions for this equilibrium are then obtained through the Alternating Direction Method of Multipliers (ADMM) algorithm. Moreover, owing to incomplete information caused by the requirements for privacy protection, we proposed a multi-agent deep reinforcement learning algorithm named MALPPO. MALPPO facilitates learning the Stackelberg Equilibrium without requiring private information from others, relying solely on past experiences. Comprehensive experimental results demonstrate that our MALPPO-based incentive mechanism outperforms baseline approaches significantly, showcasing rapid convergence and achieving the highest reward.

High-dimensional small sample size data, which may lead to singularity in computation, are becoming increasingly common in the field of pattern recognition. Moreover, it is still an open problem how to extract the most suitable low-dimensional features for the support vector machine (SVM) and simultaneously avoid singularity so as to enhance the SVM's performance. To address these problems, this article designs a novel framework that integrates the discriminative feature extraction and sparse feature selection into the support vector framework to make full use of the classifiers' characteristics to find the optimal/maximal classification margin. As such, the extracted low-dimensional features from high-dimensional data are more suitable for SVM to obtain good performance. Thus, a novel algorithm, called the maximal margin SVM (MSVM), is proposed to achieve this goal. An alternatively iterative learning strategy is adopted in MSVM to learn the optimal discriminative sparse subspace and the corresponding support vectors. The mechanism and the essence of the designed MSVM are revealed. The computational complexity and convergence are also analyzed and validated. Experimental results on some well-known databases (including breastmnist, pneumoniamnist, colon-cancer, etc.) show the great potential of MSVM against classical discriminant analysis methods and SVM-related methods, and the codes can be available on https://www.scholat.com/laizhihui.

Support vector machine (SVM), as a supervised learning method, has different kinds of varieties with significant performance. In recent years, more research focused on nonparallel SVM, where twin SVM (TWSVM) is the typical one. In order to reduce the influence of outliers, more robust distance measurements are considered in these methods, but the discriminability of the models is neglected. In this article, we propose robust manifold twin bounded SVM (RMTBSVM), which considers both robustness and discriminability. Specifically, a novel norm, that is, capped$L_{1}$-norm, is used as the distance metric for robustness, and a robust manifold regularization is added to further improve the robustness and classification performance. In addition, we also use the kernel method to extend the proposed RMTBSVM for nonlinear classification. We introduce the optimization problems of the proposed model. Subsequently, effective algorithms for both linear and nonlinear cases are proposed and proved to be convergent. Moreover, the experiments are conducted to verify the effectiveness of our model. Compared with other methods under the SVM framework, the proposed RMTBSVM shows better classification accuracy and robustness.

Syndrome-trellis codes (STCs) are commonly used in image steganographic schemes, which aim at minimizing the embedding distortion, but most distortion models cannot capture the mutual interaction of embedding modifications (MIEMs). In this article, a secure halftone image steganographic scheme based on a feature space and layer embedding is proposed. First, a feature space is constructed by a characterization method that is designed based on the statistics of 4 ×4 pixel blocks in halftone images. Upon the feature space, a generalized steganalyzer with good classification ability is proposed, which is used to measure the embedding distortion. As a result, a distortion model based on a hybrid feature space is constructed, which outperforms some state-of-the-art models. Then, as the distortion model is established on the statistics of local regions, a layer embedding strategy is proposed to reduce MIEM. It divides the host image into multiple layers according to their relative positions in 4 ×4 blocks, and the embedding procedure is executed layer by layer. In each layer, any two pixels are located at different 4 ×4 blocks in the original image, and the distortion model makes sure that the calculation of pixel distortions is independent. Between layers, the pixel distortions of the current layer are updated according to the previous embedding modifications, thus reducing the total embedding distortion. Comparisons with prior schemes demonstrate that the proposed steganographic scheme achieves high statistical security when resisting the state-of-the-art steganalysis.

Social networks are everywhere and currently transmitting very large messages. As a result, transmitting secret messages in such an environment is worth researching. However, the images used in transmitting messages are usually compressed with a JPEG compression channel, which is lossy and damages the transmitted data. Therefore, to prevent secret messages from being damaged, a robust JPEG steganography is urgently needed. In this paper, a secure robust JPEG steganographic scheme based on an autoencoder with an adaptive BCH encoding (Bose-Chaudhuri-Hocquenghem encoding) is proposed. In particular, the autoencoder is first pretrained to fit the transformation relationship between the JPEG image before and after compression by the compression channel. In addition, the BCH encoding is adaptively utilized according to the content of cover image to decrease the error rate of secret message extraction. The DCT (Discrete Cosine Transformation) coefficient adjustment based on practical JPEG channel characteristics further improves the robustness and statistical security. Comparisons with prior state-of-the-art schemes demonstrate that the proposed robust JPEG steganographic algorithm can provide a more robust performance and statistical security.

Targeting the mode-mixing problem of intrinsic time-scale decomposition (ITD) and the parameter optimization problem of least-square support vector machine (LSSVM), we propose a novel approach based on complete ensemble intrinsic time-scale decomposition (CEITD) and LSSVM optimized by the hybrid differential evolution and particle swarm optimization (HDEPSO) algorithm for the identification of the fault in a diesel engine. The approach consists mainly of three stages. First, to solve the mode-mixing problem of ITD, a novel CEITD method is proposed. Then the CEITD method is used to decompose the nonstationary vibration signal into a set of stationary proper rotation components (PRCs) and a residual signal. Second, three typical types of time-frequency features, namely singular values, PRCs energy and energy entropy, and AR model parameters, are extracted from the first several PRCs and used as the fault feature vectors. Finally, a HDEPSO algorithm is proposed for the parameter optimization of LSSVM, and the fault diagnosis results can be obtained by inputting the fault feature vectors into the HDEPSO-LSSVM classifier. Simulation and experimental results demonstrate that the proposed fault diagnosis approach can overcome the mode-mixing problem of ITD and accurately identify the fault patterns of diesel engines.

Measurements for future outdoor cellular systems at 28 GHz and 38 GHz were conducted in urban microcellular environments in New York City and Austin, Texas, respectively. Measurements in both line-of-sight and non-line-of-sight scenarios used multiple combinations of steerable transmit and receive antennas (e.g. 24.5 dBi horn antennas with 10.9° half power beamwidths at 28 GHz, 25 dBi horn antennas with 7.8° half power beamwidths at 38 GHz, and 13.3 dBi horn antennas with 24.7° half power beamwidths at 38 GHz) at different transmit antenna heights. Based on the measured data, we present path loss models suitable for the development of fifth generation (5G) standards that show the distance dependency of received power. In this paper, path loss is expressed in easy-to-use formulas as the sum of a distant dependent path loss factor, a floating intercept, and a shadowing factor that minimizes the mean square error fit to the empirical data. The new models are compared with previous models that were limited to using a close-in free space reference distance. Here, we illustrate the differences of the two modeling approaches, and show that a floating intercept model reduces the shadow factors by several dB and offers smaller path loss exponents while simultaneously providing a better fit to the empirical data. The upshot of these new path loss models is that coverage is actually better than first suggested by work in [1], [7] and [8].