Search

In this article, an optimal output synchronization solution to the $H_{\infty }$ optimization of linear discrete-time (DT) multiagent systems is investigated. Compared with current approaches, the issue of designated convergence rate is handled with system optimality, while less computation cost is required. Specifically, the internal model principle is employed to derive a cooperative regulation problem of DT systems, wherein no explicit solution to output regulation equations is needed for learning. Then, we introduce a convergence rate parameter to construct a group of auxiliary cooperative systems, based on which the zero-sum game in $H_{\infty }$ optimization is formulated. The data-efficient off-policy reinforcement learning and output-feedback technique are applied to solve the enhanced Bellman equations with a designated convergence rate. This results in an online optimal synchronization solution learning from only the input-output data along the system trajectories. It is shown that the proposed optimal synchronization protocol achieves asymptotic synchronization for the original systems with the consensus error converging to zero at a designated rate. The effectiveness of the proposed approach is verified by the simulation results.

Machine learning (ML)-based Global Navigation Satellite System (GNSS) signal recognition has been considered an efficient way to mitigate multipath error in positioning estimation. However, because of the spatial heterogeneity of nonline-of-sight (NLOS) interference, existing NLOS signal recognition methods struggle to achieve satisfactory accuracy across multiple locations. In particular, current methods achieve limited recognition accuracy on target domains containing unfamiliar environments. This study proposes a few-shot learning-based semi-supervised GNSS NLOS recognition method using cross-spatiotemporal information, which can achieve high-precision recognition for multiple target domains containing unfamiliar environments only requiring a limited number of labeled samples. First, we propose a self-attention mechanism-based cross-spatiotemporal information (SAM-CSI) module that captures the environmental characteristics used to model the spatial heterogeneity of NLOS interference. Next, to avoid the need for numerous labeled samples in training the model, we introduce a Siamese neural network-based few-shot learning (SNN-FSL) architecture to perform NLOS recognition using a similarity metric instead of label prediction. We then develop a semi-supervised NLOS recognition method by integrating the SAM-CSI module with the SNN-FSL architecture to enable high-precision recognition in target domains containing unfamiliar environments. Finally, we conducted comparison and few-shot learning experiments using real-world datasets, which demonstrated that our proposed method could achieve over 90% recognition accuracy in target domains containing unfamiliar environments after 50-shot learning. It represented an improvement of over 19% compared with existing methods.

Deep reinforcement learning (DRL) has been applied to satellite navigation and positioning applications. Its performance relies heavily on the function-approximation capability of deep neural networks. However, existing DRL models suffer from catastrophic interference, resulting in inaccurate function approximation. The sparse-coding-based DRL is an effective method to mitigating this interference, but existing methods involve the following two challenging issues: first, the value function estimation network suffers from instability problems with gradient backpropagation, including gradient explosion and gradient vanishing, second, existing methods are limited to using hand-crafted sparse regularizers that produce only static sparsity, which may be difficult to apply in various dynamic reinforcement learning (RL) environments. In this article, we propose a novel dictionary learning (DL)-structured RL model with adaptive-sparsity regularizer (ASR) that alleviates the catastrophic interference and enables accurate value function approximation, thereby improving the RL performance. To alleviate the interference and avoid the instability problems in RL, a feedforward DL-structured RL model is constructed to predict the value function without the need for gradient backpropagation. To learn data-driven sparse representations with adaptive sparsity, we propose to use the learnable sparse regularizer ASR in the model, where the key hyperparameters of ASR can be trained to be adaptive to variable RL environments. To optimize the model efficiently, the model parameters are first pretrained in the pretraining stage, with only the value weights used for value function approximation needing to be fine-tuned for actual RL applications in the control training stage. Our comparative experiments in benchmark environments demonstrate that the proposed method can outperform existing state-of-the-art sparse-coding-based RL algorithms. In terms of accumulated rewards (used to measure the quality of the learned policy), the improvement was over 63% in Cart Pole environment and nearly 10% for Puddle World. Furthermore, the proposed algorithm can maintain its relatively high performance in the presence of noise up to 20 dB.

Accurately forecasting the motion of surrounding vehicles is a crucial prerequisite for achieving safe autonomous driving (AD). Methods for trajectory prediction encompass both physics- and learning-based approaches. Despite the significant advancements made by learning-based methods in enhancing performance, ensuring that predicted trajectories are realistic, interpretable, and physically feasible remains a challenging problem. In this study, we propose a physics-based deep learning (DL) framework founded on an encoder–decoder architecture, modeling the historical motion of traffic agents effectively coupled with the surrounding environment information through attention mechanisms. Our model incorporates the vehicle dynamic model (VDM) to couple with vehicle motion and utilizes the Kalman filter to fuse scene context information for multistep accurate and feasible multimodal trajectory prediction. Our method leverages the strengths of both learning- and physics-based models. The extensive experiment results on the Lyft l5 dataset demonstrate that our model outperforms various baseline approaches in terms of metrics including prediction errors, feasibility, and explainability.

Multiview dictionary learning (DL) is attracting attention in multiview clustering due to the efficient feature learning ability. However, most existing multiview DL algorithms are facing problems in fully utilizing consistent and complementary information simultaneously in the multiview data and learning the most precise representation for multiview clustering because of gaps between views. This article proposes an efficient multiview DL algorithm for multiview clustering, which uses the partially shared DL model with a flexible ratio of shared sparse coefficients to excavate both consistency and complementarity in the multiview data. In particular, a differentiable scale-invariant function is used as the sparsity regularizer, which considers the absolute sparsity of coefficients as the $\ell _{0}$ norm regularizer but is continuous and differentiable almost everywhere. The corresponding optimization problem is solved by the proximal splitting method with extrapolation technology; moreover, the proximal operator of the differentiable scale-invariant regularizer can be derived. The synthetic experiment results demonstrate that the proposed algorithm can recover the synthetic dictionary well with reasonable convergence time costs. Multiview clustering experiments include six real-world multiview datasets, and the performances show that the proposed algorithm is not sensitive to the regularizer parameter as the other algorithms. Furthermore, an appropriate coefficient sharing ratio can help to exploit consistent information while keeping complementary information from multiview data and thus enhance performances in multiview clustering. In addition, the convergence performances show that the proposed algorithm can obtain the best performances in multiview clustering among compared algorithms and can converge faster than compared multiview algorithms mostly.

Label noise is often contained in the training data due to various human factors or measurement errors, which significantly causes a negative effect on classifiers. Despite many previous methods that have been proposed to learn robust classifiers, they are mainly based on the single-view feature. On the other hand, although existing multi-view classification methods benefit from the more comprehensive information, they rarely consider label noise. In this paper, we propose a novel label-noise robust classification model with multi-view learning to overcome these limitations. In the proposed model, not only the classifier learning but also the label-noise removal can benefit from the multi-view information. Specifically, we relax the label matrix of the basic multi-view least squares regression model, and develop a nonlinear transformation with a natural probabilistic approximation in the process of labels, which is conveniently optimized and beneficial to improve the discriminative ability of classifiers. Moreover, we preserve the intrinsic manifold structure of multi-view data on the relaxed label matrix, facilitating the process of label relaxation. For optimizing the proposed model with the nonlinear transformation, we derive a lemma about the partial derivation of the softmax related function, and develop an efficient alternating algorithm. Experimental evaluations on six real-world datasets confirm the advantages of the proposed method, compared to the related state-of-the-art methods.

Deep reinforcement learning (DRL), which highly depends on the data representation, has shown its potential in many practical decision-making problems. However, the process of acquiring representations in DRL is easily affected by interference from models, and moreover leaves unnecessary parameters, leading to control performance reduction. In this article, we propose a double sparse DRL via multilayer sparse coding and nonconvex regularized pruning. To alleviate interference in DRL, we propose a multilayer sparse-coding-structural network to obtain deep sparse representation for control in reinforcement learning. Furthermore, we employ a nonconvex log regularizer to promote strong sparsity, efficiently removing the unnecessary weights with a regularizer-based pruning scheme. Hence, a double sparse DRL algorithm is developed, which can not only learn deep sparse representation to reduce the interference but also remove redundant weights while keeping the robust performance. The experimental results in five benchmark environments of the deep $q$ network (DQN) architecture demonstrate that the proposed method with deep sparse representations from the multilayer sparse-coding structure can outperform existing sparse-coding-based DRL in control, for example, completing Mountain Car with 140.81 steps, achieving near 10% reward increase from the single-layer sparse-coding DRL algorithm, and obtaining 286.08 scores in Catcher, which are over two times the rewards of the other algorithms. Moreover, the proposed algorithm can reduce over 80% parameters while keeping performance improvements from deep sparse representations.

Direct-optimization-based dictionary learning has attracted increasing attention for improving computational efficiency. However, the existing direct optimization scheme can only be applied to limited dictionary learning problems, and it remains an open problem to prove that the whole sequence obtained by the algorithm converges to a critical point of the objective function. In this article, we propose a novel direct-optimization-based dictionary learning algorithm using the minimax concave penalty (MCP) as a sparsity regularizer that can enforce strong sparsity and obtain accurate estimation. For solving the corresponding optimization problem, we first decompose the nonconvex MCP into two convex components. Then, we employ the difference of the convex functions algorithm and the nonconvex proximal-splitting algorithm to process the resulting subproblems. Thus, the direct optimization approach can be extended to a broader class of dictionary learning problems, even if the sparsity regularizer is nonconvex. In addition, the convergence guarantee for the proposed algorithm can be theoretically proven. Our numerical simulations demonstrate that the proposed algorithm has good convergence performances in different cases and robust dictionary-recovery capabilities. When applied to sparse approximations, the proposed approach can obtain sparser and less error estimation than the different sparsity regularizers in existing methods. In addition, the proposed algorithm has robustness in image denoising and key-frame extraction.

Despite the ability of 3D convolutional methods to extract spatio-temporal information simultaneously, they also increase parameter redundancy and computational and storage costs. Previous work that has utilized the 2D convolution method has approached the problem in one of two ways: either using the entire body sequence as input to extract global features or dividing the body sequence into several parts to extract local features. However, global information tends to overlook detailed information specific to each body part, while local information fails to capture relationships between local regions. Therefore, this study proposes a new framework for constructing spatio-temporal representations, which involves extracting and fusing features in a novel manner. To achieve this, we introduce the multi-feature extraction-fusion (MFEF) module, which includes two branches: each branch extracts global features or local features individually, after which they are fused using multiple strategies. Additionally, as gait is a periodic action and different body parts contribute unequally to recognition during each cycle, we propose the periodic temporal feature modeling (PTFM) module, which extracts temporal features from adjacent frame parts during the complete gait cycle, based on the fused features. Furthermore, to capture fine-grained information specific to each body part, our framework utilizes multiple parallel PTFMs to correspond with each body part. We conducted a comprehensive experimental study on the widely used public dataset CASIA-B. Results indicate that the proposed approach achieved an average rank-1 accuracy of 97.2% in normal walking conditions, 92.3% while carrying a bag during walking, and 80.5% while wearing a jacket during walking.

Key frame extraction based on sparse coding can reduce the redundancy of continuous frames and concisely express the entire video. However, how to develop a key frame extraction algorithm that can automatically extract a few frames with a low reconstruction error remains a challenge. In this paper, we propose a novel model of structured sparse-coding-based key frame extraction, wherein a nonconvex group log-regularizer is used with strong sparsity and a low reconstruction error. To automatically extract key frames, a decomposition scheme is designed to separate the sparse coefficient matrix by rows. The rows enforced by the nonconvex group log-regularizer become zero or nonzero, leading to the learning of the structured sparse coefficient matrix. To solve the nonconvex problems due to the log-regularizer, the difference of convex algorithm (DCA) is employed to decompose the log-regularizer into the difference of two convex functions related to the $I_{1}$ norm, which can be directly obtained through the proximal operator. Therefore, an efficient structured sparse coding algorithm with the group log-regularizer for key frame extraction is developed, which can automatically extract a few frames directly from the video to represent the entire video with a low reconstruction error. Experimental results demonstrate that the proposed algorithm can extract more accurate key frames from most SumMe videos compared to the state-of-the-art methods. Furthermore, the proposed algorithm can obtain a higher compression with a nearly 18% increase compared to sparse modeling representation selection (SMRS) and an 8% increase compared to SC-det on the VSUMM dataset.

To reasonably utilize the coal direct liquefaction residue (DLR), contrasting research on the co-pyrolysis between different low-rank coals and DLR was investigated using a TGA coupled with an FT-IR spectrophotometer and a fixed-bed reactor. GC–MS, FTIR, and XRD were used to explore the reaction mechanisms of the various co-pyrolysis processes. Based on the TGA results, it was confirmed that the tetrahydrofuran insoluble fraction of DLR helped to catalyze the conversion reaction of lignite. Also, the addition of DLR improved the yield of tar in the fixed-bed, with altering the composition of the tar. Moreover, a kinetic analysis during the co-pyrolysis was conducted using a distributed activation energy model. The co-pyrolysis reactions showed an approximate double-Gaussian distribution.

Nonlinear forced vibrations and natural frequency of sandwich functionally graded material doubly curved shallow shell with a rectangular base are investigated. The sandwich functionally graded material (FGM) doubly curved shell is subjected to a harmonic point load at centre. The sandwich doubly curved shell with homogeneous face sheets and FGM face sheets is considered respectively when the natural frequencies are studied. Reddy’s third order shear deformation theory is expanded in which stretching effects in thickness are considered by introducing the secant function. Hamilton’s principle and von-Karman type nonlinear geometric equation are applied to obtain partial differential equation of the FGM sandwich doubly curved shell. Comparative studies with other shear deformation theories are carried out to validate the present formulation. Navier method is used to discuss the natural vibration frequencies of the FGM sandwich doubly curved shell. Numerical simulation is applied to demonstrate the nonlinear dynamic responses of the FGM sandwich doubly curved shell. Multiple periods, quasi-period and chaos are detected for the dynamic system for different core thickness.

We combine the active surface of polydopamine (PDA) with the biomimetic mineralization of CaCO3to obtain the macroscopically continuous CaCO3films under mild conditions. In this approach, the organic matrices were adhesive PDA coatings, which were dip-coated on silicon wafers by the self-polymerization of dopamine in an alkaline aqueous solution (pH 8.5). The inorganic layers were CaCO3films, which were formed in a CaCl2solution in the presence of poly(acrylic acid) (PAA) via a CO2diffusion method. During the biomimetic mineralization, amorphous calcium carbonate (ACC) was formed on PDA matrices with the help of PAA, which, subsequently, was transformed into a flat continuous calcite film on the PDA matrices. As the mineralization time increased, a new layer of CaCO3crystals was formed over the calcite and, as a result, led to continuous CaCO3films with rough surfaces. The thicknesses of CaCO3films can be controlled by tuning the mineralization time. Our approach may provide a simple, yet efficient way for the preparation of macroscopically continuous organic–inorganic composite CaCO3films under mild conditions. Moreover, superhydrophobic surfaces can be successfully achieved via a hydrophobic modification of the rough CaCO3films, which make them suitable candidates for a variety of superhydrophobic applications, such as self-cleaning surfaces or anticorrosion, antiadhesive coatings.

Transfer learning is an important technology in addressing the problem that labeled data in a target domain are difficult to collect using extensive labeled data from the source domain. Recently, an algorithm named graph co-regularized transfer learning (GTL) has shown a competitive performance in transfer learning. However, its convergence is affected by the used approximate scheme, degenerating learned results. In this paper, after analyzing convergence conditions, we propose a novel update rule using the multiplicative update rule and develop a new algorithm named improved GTL (IGTL) with a strict convergence guarantee. Moreover, to prove the convergence of our method, we design a special auxiliary function whose value is intimately related to that of the objective function. Finally, the experimental results on the synthetic dataset and two real-world datasets confirm that the proposed IGTL is convergent and performs better than the compared methods.

The beneficial effects of a new recombinant lipase (Rhizopus chinensislipase [RCL]) and transglutaminase (TG) were investigated on frozen dough systems and their breadmaking quality. Rheological properties and microstructure of doughs were measured using a dynamic rheometer, rheofermentometer F3, and scanning electron microscopy (SEM). Measurements of viscoelastic properties showed that both G' and G? of dough containing RCL and TG were greater than those of the control after 35 days of frozen storage. The SEM micrographs showed that dough containing RCL and TG had the most starch granules embedded in or attached to the gluten network, and the gluten seemed more powerful and resilient than for the control dough after 35 days of frozen storage. Results of the gas production and dough development tests indicated that RCL and TG improved the rheofermentative characteristics of frozen dough. RCL and TG could improve water-holding capacity and significantly increase the glycerol content of the control dough. Image analyses showed that bread crumbs containing RCL and TG had a more open network and uniform crumb structure, which resulted in higher specific volume. This combination also yielded a product with higher sensory scores for test breads.