Article
Automation & Control Systems
Yan-Hui Lin, Liang Chang
Summary: This article proposes a novel online transfer learning framework for a regression task, which addresses the issue of different distributions and time-varying distributions of data through offline and online stages of learning and updating strategies, and utilizes the knowledge from the source domain through an ensemble approach.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Computer Science, Artificial Intelligence
Aleksandra Knapinska, Piotr Lechowicz, Weronika Wegier, Krzysztof Walkowiak
Summary: This study proposes a chunk-based ensemble learning method for long-term prediction of various network traffic types, which does not require large volumes of training data and is resilient to traffic characteristic changes. Experimental results show that the proposed method outperforms reference methods in prediction errors, providing better network resource adjustment recommendations for network operators.
APPLIED SOFT COMPUTING
(2022)
Article
Thermodynamics
Huaiping Jin, Yunlong Li, Bin Wang, Biao Yang, Huaikang Jin, Yundong Cao
Summary: Wind power has become an important part of clean energy. This study proposes an adaptive wind power forecasting method called selective ensemble of offline global and online local learning (SEOGOL). The method enhances the diversity of base models through a multi-modal perturbation mechanism and achieves adaptive fusion of individual models according to the Bayesian rule. The effectiveness and superiority of the proposed SEOGOL method are verified using an actual wind power data set.
ENERGY CONVERSION AND MANAGEMENT
(2022)
Article
Engineering, Mechanical
Zheng Zhou, Tianfu Li, Zhibin Zhao, Chuang Sun, Xuefeng Chen, Ruqiang Yan, Jide Jia
Summary: This paper proposes a dynamic governing network (DGN) to address the issue of remaining useful life (RUL) prediction, by formulating it as a time-varying trajectory modeling problem and utilizing a discretized ordinary differential equation (ODE) parameterized by neural networks. The time-varying network architecture of the DGN is dynamically determined by a deep reinforcement learning algorithm. Experimental results demonstrate that the proposed DGN can capture underlying dynamics from observation series and achieve state-of-the-art RUL prediction performance.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Computer Science, Information Systems
Shao-Chun Wen, Cheng-Hsiung Yang
Summary: The paper introduces a framework for predicting the value of time series for nonlinear systems, including four nonlinear systems and three learning parts. K-means method and MAE are used to evaluate the prediction accuracy.
INFORMATION SCIENCES
(2021)
Article
Environmental Sciences
Christopher Krapu, Mark Borsuk
Summary: This study proposes using Hamiltonian Monte Carlo for Bayesian inference for time-varying parameters (TVP) in hydrology models. The results from simulation experiments and analysis of real-world data suggest that gradient-based MCMC is a flexible and reliable approach for analyzing hydrology models with dynamic parameter sets.
WATER RESOURCES RESEARCH
(2022)
Article
Environmental Sciences
Hojjatollah Mahboobi, Alireza Shakiba, Babak Mirbagheri
Summary: Groundwater is a crucial water resource that is increasingly contaminated. This study aims to improve the spatial accuracy of predicting groundwater nitrate concentration through integrating machine learning models using a local approach. The findings demonstrate that the ensemble of ML models using geographically weighted regression achieves the highest predictive performance.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2023)
Article
Forestry
Yuling Chen, Chen Dong, Baoguo Wu
Summary: This study explores the application of ensemble learning in crown profile modeling and prediction in forest science. The results show that ensemble learning has better performance and efficiency, and is effective in improving model accuracy.
Article
Automation & Control Systems
Xuanyu Cao, Junshan Zhang, H. Vincent Poor
Summary: This article studies online stochastic optimization with time-varying distributions, presenting algorithms that consider dynamic optimal points to define regret and performance benchmark. The proposed algorithms demonstrate sublinear regret as long as the drift of the optima is sublinear, indicating effective performance in constrained stochastic optimization scenarios.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2021)
Article
Engineering, Mechanical
ChengJiu Zhu, HaiDong Yang, YaJun Fan, Bi Fan, KangKang Xu
Summary: A novel online spatiotemporal modeling method is proposed to model time-varying distributed parameter systems using Kernel-based Multilayer Extreme Learning Machine. The method creates a deep network by stacking multiple Kernel-based Extreme Learning Machine Autoencoders and one original Extreme Learning Machine Autoencoder, transforming the spatiotemporal output of time-varying DPSs into low-dimensional time coefficients and reconstructing the spatiotemporal dynamics using Kernel-based Extreme Learning Machine.
NONLINEAR DYNAMICS
(2022)
Article
Engineering, Mechanical
Yuquan Meng, Chenhui Shao
Summary: This paper proposes a novel hierarchical physics-informed ensemble learning (PIEL) framework for accurate online prediction of ultrasonic metal welding (UMW) joint strength. The framework decomposes the variability of joint strength into a physics-informed global trend and a data-driven residual, enabling the establishment of hierarchical prediction models. A highly efficient feature extraction procedure based on discrete wavelet transformation (DWT) is developed to automatically extract key low-dimensional information from high-dimensional sensing signals. The effectiveness of the PIEL framework is demonstrated through two real-world case studies, achieving superior prediction accuracy compared to state-of-the-art baseline methods and showing excellent modeling robustness and data efficiency by integrating physical knowledge.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Automation & Control Systems
Nishtha Hooda, Jasgurpreet Singh Chohan, Ruchika Gupta, Raman Kumar
Summary: In this study, artificial intelligence systems are integrated with mechanical systems to reduce manufacturing time and cost of products. The random forest machine learning model is used to predict and validate the optimum deposition angle in Fused Deposition Modeling, achieving a prediction accuracy of 94.57%, significantly better than other methods. This robust model efficiently predicts the optimum deposition angle for any geometry, enhancing the applicability of digitally manufactured products.
Article
Geosciences, Multidisciplinary
Huajin Zhang, Shunchuan Wu, Xiaoqiang Zhang, Longqiang Han, Zhongxin Zhang
Summary: In order to evaluate slope stability quickly, accurately, and reliably, a slope stability prediction method based on margin distance minimization selective ensemble (MDMSE) is proposed. This method objectively evaluates slope stability using basic geometric and geological factors, overcoming the disadvantages of traditional machine learning models in terms of difficult selection and high misjudgment risk. Experimental results show that the MDMSE prediction model has better generalization ability, recognition accuracy, and identification speed compared to other models. The MDMSE prediction model is suitable for slope stability prediction and analysis and has practical engineering references.
Article
Automation & Control Systems
Mingchuan Zhang, Bowei Hao, Quanbo Ge, Junlong Zhu, Ruijuan Zheng, Qingtao Wu
Summary: D-AdaBound is a distributed adaptive subgradient algorithm that dynamically clips learning rates to achieve good performance. It has a reasonable regret bound under convex objective functions. Experimental results demonstrate performance improvement of D-AdaBound relative to existing distributed online learning algorithms on different datasets.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Engineering, Civil
Tianli Guo, Songbai Song, Vijay P. Singh, Ting Wei, Te Zhang, Xin Liu
Summary: This study developed a time-varying stepwise decomposition ensemble framework for nonsta-tionary and nonlinear streamflow series, along with an optimization strategy combining a two-stage calibration strategy with a particle swarm optimization algorithm. The results showed that the time-varying decomposition ensemble models were superior to the single models, and the TSC-PSO-PSO optimization strategy outperformed other optimization strategies. The TV-VMD-SVM model based on the TSC-PSO-PSO optimization strategy had the best streamflow forecasting performance.
JOURNAL OF HYDROLOGY
(2023)
Article
Telecommunications
Zhaohui Huang, Dongxuan He, Jiaxuan Chen, Zhaocheng Wang, Sheng Chen
Summary: This article proposes an improved autoencoder to enhance the robustness of Terahertz wireless communication against hybrid distortions. It encodes and decodes transmitted symbols and utilizes a fitting network to approximate the channel characteristics. Simulation results demonstrate that the proposed method can recover symbols and improve demodulation performance significantly under severe hybrid distortions.
CHINA COMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Xin Sun, Changrui Chen, Xiaorui Wang, Junyu Dong, Huiyu Zhou, Sheng Chen
Summary: This study proposes a lightweight neural network approach based on Gaussian dynamic convolution (GDC) for single-image segmentation. The GDC method efficiently aggregates contextual information to achieve superior segmentation results, and demonstrates potential and generality in common semantic segmentation.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Tong Liu, Sheng Chen, Shan Liang, Shaojun Gan, Chris J. Harris
Summary: This article proposes a multi-output selective ensemble regression (SER) method for online identification of multi-output nonlinear time-varying industrial processes. The method achieves low online computational complexity and high prediction accuracy through adaptive local learning and optimization of the combining weights of the selected local multi-output models based on a probability metric.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Automation & Control Systems
Chao Tan, Sheng Chen, Genlin Ji, Xin Geng
Summary: This study proposes a multi-label distribution learning algorithm based on multi-output regression and manifold learning, called MDLRML. By utilizing the smooth and similar information provided by manifold learning and label distribution learning, the algorithm links the manifolds in two spaces, enhancing classification accuracy and efficiency.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Computer Science, Artificial Intelligence
Chao Tan, Sheng Chen, Genlin Ji, Xin Geng
Summary: In this paper, we propose a novel probabilistic label enhancement algorithm (PLEA) to address the label distribution learning problem in multi-label classification. By utilizing manifold learning, we can improve the accuracy of estimated label distributions associated with feature extraction and reduce computational complexity. Experimental results demonstrate the advantages of our proposed method in terms of accuracy and runtime performance.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Engineering, Electrical & Electronic
Yinxiao Zhuo, Ziyuan Sha, Zhaocheng Wang, Sheng Chen
Summary: In this paper, two angular domain separation based multi-beam training schemes were proposed to enable simultaneous training under single data stream constraint, with optimized non-orthogonal codebooks for improved beamforming gains. Simulation results demonstrate the effectiveness and superior performance of the proposed schemes over existing state-of-the-art technologies.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Chao Tan, Sheng Chen, Xin Geng, Genlin Ji
Summary: In this paper, a novel label distribution manifold learning (LDML) method is proposed for accurately solving the multilabel distribution learning problem. Through manifold learning and multi-output kernel regression, accurate label distributions can be estimated and an enhanced maximum entropy model is formed. Experimental results demonstrate the advantages of the proposed LDML method in terms of learning accuracy.
PATTERN RECOGNITION
(2023)
Article
Engineering, Electrical & Electronic
Ke Ma, Shouliang Du, Haoming Zou, Wenqiang Tian, Zhaocheng Wang, Sheng Chen
Summary: Motivated by the inter-BS channel dependence, this paper proposes a method to fuse sub-6 GHz channel information and mmWave low-overhead measurement to predict the optimal mmWave beam in HetNets, reducing overhead and achieving high prediction accuracy using deep learning.
IEEE TRANSACTIONS ON COMMUNICATIONS
(2023)
Article
Engineering, Electrical & Electronic
Jiankang Zhang, Dong Liu, Sheng Chen, Soon Xin Ng, Robert G. Maunder, Lajos Hanzo
Summary: This article proposes a discrete ε multi-objective genetic algorithm (ε-DMOGA) to optimize the end-to-end latency, end-to-end spectral efficiency (SE), and path expiration time (PET) in aeronautical ad-hoc networks. It incorporates a distance-based adaptive coding and modulation (ACM) scheme and queueing delay into the optimization metric.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Telecommunications
Fuwang Dong, Wei Wang, Xin Li, Fan Liu, Sheng Chen, Lajos Hanzo
Summary: This paper presents an advanced wireless solution based on the dual-functional radar and communication (DFRC) technique, and designs sophisticated beamforming schemes by considering physical layer security (PLS). It is shown that properly designed radar waveforms can serve as artificial noise for eavesdropping prevention and provide increased design degrees of freedom. A non-convex optimization problem is formulated to strike a balance among conflicting objectives, and a semidefinite relaxation (SDR)-based algorithm is proposed to find globally optimal solutions. A robust beamforming method is also developed to handle imperfect channel state information (CSI). Simulation results confirm the theoretical findings and demonstrate the superiority of the proposed methods.
IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING
(2023)
Article
Computer Science, Artificial Intelligence
Chao Tan, Sheng Chen, Xin Geng, Genlin Ji
Summary: This study presents a label enhancement model to address the multi-label learning problem, improving the ability of label recognition through incremental subspace learning. Experimental results demonstrate its superior predictive performance compared to other methods.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Ting Zhang, Muhammad Waqas, Yu Fang, Zhaoying Liu, Zahid Halim, Yujian Li, Sheng Chen
Summary: This paper proposes a weakly-supervised butterfly detection model based on a saliency map (WBD-SM) to enhance the accuracy of butterfly detection in the ecological environment. Experimental results show that WBD-SM achieves higher recognition accuracy than VGG16 under different division ratios.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Yuxin Meng, Eric Rigall, Xueen Chen, Feng Gao, Junyu Dong, Sheng Chen
Summary: This article proposes a framework combining generative adversarial networks (GANs) and numerical models for sea subsurface temperature prediction. By using GANs to learn the simplified physics between surface temperature and target subsurface temperature, and calibrating the parameters with observational data, the framework improves prediction performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Tong Liu, Zeyue Tian, Sheng Chen, Kai Wang, Chris J. Harris
Summary: This article proposes a deep cascade GRBF network for online modeling and prediction of high-dimensional processes with nonstationary characteristics. By integrating feature learning and online adaptation, the proposed method improves prediction accuracy and computational efficiency.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Engineering, Civil
Akhtar Badshah, Muhammad Waqas, Fazal Muhammad, Ghulam Abbas, Ziaul Haq Abbas, Shehzad Ashraf Chaudhry, Sheng Chen
Summary: This study proposes an anonymous authenticated key exchange mechanism (AAKE-BIVT) for the smart transportation Internet of Vehicles (IoVs) supported by blockchain. AAKE-BIVT securely transmits traffic information and generates transactions through mutual authentication and key agreement. The study reveals that AAKE-BIVT is resistant to security attacks and outperforms existing techniques in terms of security, functionality, communication, and computation.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Rui Lv, Dingheng Wang, Jiangbin Zheng, Zhao-Xu Yang
Summary: In this paper, the authors investigate tensor decomposition for neural network compression. They analyze the convergence and precision of tensor mapping theory, validate the rationality of tensor mapping and its superiority over traditional tensor approximation based on the Lottery Ticket Hypothesis. They propose an efficient method called 3D-KCPNet to compress 3D convolutional neural networks using the Kronecker canonical polyadic (KCP) tensor decomposition. Experimental results show that 3D-KCPNet achieves higher accuracy compared to the original baseline model and the corresponding tensor approximation model.
Article
Computer Science, Artificial Intelligence
Xiangkun He, Zhongxu Hu, Haohan Yang, Chen Lv
Summary: In this paper, a novel constrained multi-objective reinforcement learning algorithm is proposed for personalized end-to-end robotic control with continuous actions. The approach trains a single model using constraint design and a comprehensive index to achieve optimal policies based on user-specified preferences.
Article
Computer Science, Artificial Intelligence
Zhijian Zhuo, Bilian Chen, Shenbao Yu, Langcai Cao
Summary: In this paper, a novel method called Expansion with Contraction Method for Overlapping Community Detection (ECOCD) is proposed, which utilizes non-negative matrix factorization to obtain disjoint communities and applies expansion and contraction processes to adjust the degree of overlap. ECOCD is applicable to various networks with different properties and achieves high-quality overlapping community detection.
Article
Computer Science, Artificial Intelligence
Yizhe Zhu, Chunhui Zhang, Jialin Gao, Xin Sun, Zihan Rui, Xi Zhou
Summary: In this work, the authors propose a Contrastive Spatio-Temporal Distilling (CSTD) approach to improve the detection of high-compressed deepfake videos. The approach leverages spatial-frequency cues and temporal-contrastive alignment to fully exploit spatiotemporal inconsistency information.
Review
Computer Science, Artificial Intelligence
Laijin Meng, Xinghao Jiang, Tanfeng Sun
Summary: This paper provides a review of coverless steganographic algorithms, including the development process, known contributions, and general issues in image and video algorithms. It also discusses the security of coverless steganography from theoretical analysis to actual investigation for the first time.
Article
Computer Science, Artificial Intelligence
Yajie Bao, Tianwei Xing, Xun Chen
Summary: Visual question answering requires processing multi-modal information and effective reasoning. Neural-symbolic learning is a promising method, but current approaches lack uncertainty handling and can only provide a single answer. To address this, we propose a confidence based neural-symbolic approach that evaluates NN inferences and conducts reasoning based on confidence.
Article
Computer Science, Artificial Intelligence
Anh H. Vo, Bao T. Nguyen
Summary: Interior style classification is an interesting problem with potential applications in both commercial and academic domains. This project proposes a method named ISC-DeIT, which combines data-efficient image transformer architectures and knowledge distillation, to address the interior style classification problem. Experimental results demonstrate a significant improvement in predictive accuracy compared to other state-of-the-art methods.
Article
Computer Science, Artificial Intelligence
Shashank Kotyan, Danilo Vasconcellos Vargas
Summary: This article introduces a novel augmentation technique called Dynamic Scanning Augmentation to improve the accuracy and robustness of Vision Transformer (ViT). The technique leverages dynamic input sequences to adaptively focus on different patches, resulting in significant changes in ViT's attention mechanism. Experimental results demonstrate that Dynamic Scanning Augmentation outperforms ViT in terms of both robustness to adversarial attacks and accuracy against natural images.
Article
Computer Science, Artificial Intelligence
Hiba Alqasir, Damien Muselet, Christophe Ducottet
Summary: The article proposes a solution to improve the learning process of a classification network by providing shape priors, reducing the need for annotated data. The solution is tested on cross-domain digit classification tasks and a video surveillance application.
Article
Computer Science, Artificial Intelligence
Dexiu Ma, Mei Liu, Mingsheng Shang
Summary: This paper proposes a method using neural dynamics solvers to solve infinity-norm optimization problems. Two improved solvers are constructed and their effectiveness and superiority are demonstrated through theoretical analysis and simulation experiments.
Article
Computer Science, Artificial Intelligence
Francesco Gregoretti, Giovanni Pezzulo, Domenico Maisto
Summary: Active Inference is a computational framework that uses probabilistic inference and variational free energy minimization to describe perception, planning, and action. cpp-AIF is a header-only C++ library that provides a powerful tool for implementing Active Inference for Partially Observable Markov Decision Processes through multi-core computing. It is cross-platform and improves performance, memory management, and usability compared to existing software.
Article
Computer Science, Artificial Intelligence
Zelin Ying, Dawei Cheng, Cen Chen, Xiang Li, Peng Zhu, Yifeng Luo, Yuqi Liang
Summary: This paper proposes a novel stock market trends prediction framework called SMART, which includes a self-supervised stock technical data sequence embedding model S3E. By training with multiple self-supervised auxiliary tasks, the model encodes stock technical data sequences into embeddings and uses the learned sequence embeddings for predicting stock market trends. Extensive experiments on China A-Shares market and NASDAQ market prove the high effectiveness of our model in stock market trends prediction, and its effectiveness is further validated in real-world applications in a leading financial service provider in China.
Article
Computer Science, Artificial Intelligence
Hao Li, Hao Jiang, Dongsheng Ye, Qiang Wang, Liang Du, Yuanyuan Zeng, Liu Yuan, Yingxue Wang, C. Chen
Summary: DHGAT1, a dynamic hyperbolic graph attention network, utilizes hyperbolic metric properties to embed dynamic graphs. It employs a spatiotemporal self-attention mechanism and weighted node representations, resulting in excellent performance in link prediction tasks.
Article
Computer Science, Artificial Intelligence
Jiehui Huang, Zhenchao Tang, Xuedong He, Jun Zhou, Defeng Zhou, Calvin Yu-Chian Chen
Summary: This study proposes a progressive learning multi-scale feature blending model for image deraining tasks. The model utilizes detail dilation and texture extraction to improve the restoration of rainy images. Experimental results show that the model achieves near state-of-the-art performance in rain removal tasks and exhibits better rain removal realism.
Article
Computer Science, Artificial Intelligence
Lizhi Liu, Zilin Gao, Yinhe Wang, Yongfu Li
Summary: This paper proposes a novel discrete-time interconnected model for depicting complex dynamical networks. The model consists of nodes and edges subsystems, which consider the dynamic characteristic of both nodes and edges. By designing control strategies and coupling modes, the stabilization and synchronization of the network are achieved. Simulation results demonstrate the effectiveness of the proposed methods.