Article
Multidisciplinary Sciences
Mingwei Fan, Jianhong Chen, Zuanjia Xie, Haibin Ouyang, Steven Li, Liqun Gao
Summary: In this paper, an improved multi-objective differential evolution algorithm (MOEA/D/DEM) based on a decomposition strategy is proposed to enhance the search performance for practical multi-objective nutrition decision problems. The algorithm utilizes a neighborhood intimacy factor and a new Gaussian mutation strategy to improve diversity and local search ability. Experimental results show that the proposed algorithm achieves better search capability and obtains competitive results compared to other multi-objective algorithms.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Jing Liang, Kangjia Qiao, Caitong Yue, Kunjie Yu, Boyang Qu, Ruohao Xu, Zhimeng Li, Yi Hu
Summary: This paper proposes a differential evolution algorithm based on clustering technique and elite selection mechanism to solve Multimodal Multi-objective Optimization Problems (MMOPs). The algorithm calculates comprehensive crowding degree and introduces elite selection mechanism to generate a well-distributed population, resulting in superior performance compared to other commonly used algorithms, as shown in extensive experiments on CEC'2019 benchmark functions.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Engineering, Chemical
Samira Ghorbanpour, Yuwei Jin, Sekyung Han
Summary: An adaptive Grid-based Multi-Objective Differential Evolution algorithm is proposed in this paper to address multi-objective optimization, aiming to improve algorithm performance by performing mutation strategy in a grid environment and considering performance metrics.
Article
Computer Science, Artificial Intelligence
Yong Wang, Zhen Liu, Gai-Ge Wang
Summary: Recently, multimodal multi-objective problem (MMOP) has attracted significant attention in the field of multi-objective optimization problems. The proposed algorithm addresses the issue of finding all equivalent Pareto sets in MMOP by introducing a modified maximum extension distance (MMED) indicator and implementing two-stage and novel mutation strategies. Additionally, a MMED-based environmental selection strategy improves the overall performance of the population.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Automation & Control Systems
Vikas Palakonda, Jae-Mo Kang
Summary: This article proposes a preference-inspired differential evolution algorithm for multi and many-objective optimization, which effectively deals with a wide range of problems. The algorithm generates individuals with good convergence and distribution properties by utilizing a preference-inspired mutation operator and determining local knee points based on a clustering method. Experimental results demonstrate its superior performance compared to eight state-of-the-art algorithms on 35 benchmark problems.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Wu Deng, Junjie Xu, Yingjie Song, Huimin Zhao
Summary: The paper proposes a novel improved DE algorithm WMSDE, which combines wavelet basis function to optimize mutation strategy and control parameters, accelerate convergence, improve search quality, and address issues like local optima and stagnation. Experimental results demonstrate the superior performance of WMSDE on 11 benchmark functions, as well as its effectiveness in solving real-world airport gate assignment problems.
APPLIED SOFT COMPUTING
(2021)
Article
Multidisciplinary Sciences
Tassawar Ali, Hikmat Ullah Khan, Tasswar Iqbal, Fawaz Khaled Alarfaj, Abdullah Mohammad Alomair, Naif Almusallam
Summary: Differential evolution is an evolutionary algorithm that balances exploration and exploitation to find the optimal genes for an objective function. To address the challenge of finding this balance, a clustering-based mutation strategy called Agglomerative Best Cluster Differential Evolution (ABCDE) is proposed. ABCDE efficiently converges without getting trapped in local optima by clustering the population and avoiding poor-quality genes through adaptive crossover rate. ABCDE outperforms classical mutation strategies and random neighborhood mutation strategy in generating a population where the difference between the values of the trial vector and objective vector is less than 1% for some benchmark functions. The optimal and fast convergence of differential evolution has potential applications in weight optimization of artificial neural networks and stochastic/time-constrained environments like cloud computing.
Article
Computer Science, Artificial Intelligence
Liping Wang, Wei Yu, Feiyue Qiu, Yu Ren, Jiafeng Lu, Pan Fu
Summary: The paper introduces a neighbor selection strategy and dynamic preference allocation strategy to improve the solution method for multi-objective optimization problems. Experimental results demonstrate the effectiveness of the improved algorithm in tackling most multi-objective problems.
Article
Multidisciplinary Sciences
Muhammad Hassan Baig, Qamar Abbas, Jamil Ahmad, Khalid Mahmood, Sultan Alfarhood, Mejdl Safran, Imran Ashraf
Summary: Symmetry plays an important role in solving differential equations using the differential evolution (DE) algorithm. This study introduces a new mutation strategy and algorithm, DE/Neighbor/2 and IRNDE, to improve the convergence speed and performance. Experimental results show that DE/Neighbor/2 and IRNDE have better and faster convergence compared to DE/Neighbor/1 and RNDE.
Article
Computer Science, Artificial Intelligence
Xiaobo Li, Qiyong Fu, Qi Li, Weiping Ding, Feilong Lin, Zhonglong Zheng
Summary: Feature selection is a multi-objective problem that aims to choose a subset of features with minimal feature-feature correlation and maximum feature-class correlation. Grey wolf optimization mimics the hunting mechanism of grey wolves but can face local optimization in multi-objective problems. To address this, a novel multi-objective binary grey wolf optimization algorithm called MOBGWO-GMS is proposed, which utilizes a guided mutation strategy (GMS). Experimental results show that the proposed approach outperforms other algorithms in terms of the optimal trade-off between fitness evaluation criteria and the ability to escape local optima.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Xiaobing Yu, Wenguan Luo, WangYing Xu, ChenLiang Li
Summary: This study addresses the issue of selecting feasible and infeasible solutions in Constrained Multi-objective Optimization Problems (CMOPs) by developing a constrained multi-objective Differential Evolution (DE) algorithm. The experiments demonstrate that the algorithm can find well-distributed Pareto front and achieve superior performance indicator results.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Yupeng Han, Hu Peng, Changrong Mei, Lianglin Cao, Changshou Deng, Hui Wang, Zhijian Wu
Summary: This paper proposes a new multistrategy multiobjective differential evolutionary algorithm, RLMMDE, to solve the exploration and exploitation dilemma in multiobjective optimization problems (MOPs). The algorithm utilizes a multistrategy and multicrossover DE optimizer, an adaptive reference point activation mechanism based on RL, and a reference point adaptation method. Experimental results show that RLMMDE outperforms some advanced MOEAs on benchmark test suites and practical mixed-variable optimization problems.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Mathematical & Computational Biology
Shihao Yuan, Hong Zhao, Jing Liu, Binjie Song
Summary: This paper proposes a differential evolution algorithm based on self-organizing maps and dynamic selection strategy for solving multimodal optimization problems. The algorithm improves the performance of differential evolution in solving multimodal optimization problems by reasonably dividing the population, expanding the search space, and balancing exploration and exploitation.
MATHEMATICAL BIOSCIENCES AND ENGINEERING
(2022)
Review
Computer Science, Interdisciplinary Applications
Sanjoy Chakraborty, Apu Kumar Saha, Absalom E. Ezugwu, Jeffrey O. Agushaka, Raed Abu Zitar, Laith Abualigah
Summary: This article presents the recent developments of differential evolution algorithm, including parameter adaptations, parameter settings and mutation strategies, hybridizations, and multi-objective variants. It also summarizes the applications of differential evolution algorithm and its variants in image processing.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Erping Song Song, Hecheng Li
Summary: This paper proposes a new hybrid differential evolution algorithm HMODE/D for solving multi-objective optimization problems. Through optimizing local optimal selection, crossover operator design, external archive setting, and parameter self-adaptive adjustment, the algorithm achieves superior performance in solving multi-objective optimization problems.
CONNECTION SCIENCE
(2022)
Article
Instruments & Instrumentation
Min-Hao Wu, Ting-Cheng Chang, Haishan Chen, Zhenlun Yang, ShiMing Liu
Summary: This research proposes an optimized method for hiding and extracting information from image data based on the histogram shift method. Through testing five different prediction methods, it was found that the chessboard-type method performs better in image processing. The optimized prediction method shows better performance in image transmission and provides a basis for further application development.
SENSORS AND MATERIALS
(2022)
Article
Medicine, Research & Experimental
Yongli Chen, Shanwu Dong, Lin Tian, Haishan Chen, Jing Chen, Chunzhi He
Summary: This study revealed that the combination treatment of azithromycin and methylprednisolone can partially inhibit the inflammatory response, cell viability loss, and promote apoptosis induced by Mycoplasma pneumoniae infection by regulating the miR-499a-5p/STAT3 axis.
EXPERIMENTAL AND THERAPEUTIC MEDICINE
(2022)
Article
Endocrinology & Metabolism
Guo-bao Hong, Xiao-fei Shao, Jia-min Li, Qin Zhou, Xiao-Su Ke, Pei-Chun Gao, Xiao-Lin Li, Jing Ning, Hai-Shan Chen, Hua Xiao, Chong-Xiang Xiong, Hequn Zou
Summary: This study found that RBP4 is strongly associated with hyperuricemia, and its predictive value was higher than that of traditional predictors. The optimum predictive model for hyperuricemia in the general population included RBP4, sex, body mass index, serum creatinine, high-sensitivity C-reactive protein, fasting blood glucose, insulin, and alcohol consumption.
FRONTIERS IN ENDOCRINOLOGY
(2022)
Article
Environmental Sciences
Haishan Chen, Xiaoping Meng, Dianlei Liu, Wei Wang, Xiaodong Xing, Zhiyong Zhang, Chen Dong
Summary: This article presents an innovative closed-loop biosensor for online monitoring of total organic carbon (TOC) in public swimming pools (PSPs). The digital and real-time simulations demonstrate that the closed-loop MFC control system can accurately track the variation in TOC concentration in the water, providing the desired dynamic response. The proposed principle and method can also be applied for online monitoring of other substances in water and lay a theoretical foundation for MFC-based online monitoring in an aquatic environment.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2022)
Article
Meteorology & Atmospheric Sciences
Jie Cao, Qin Xu, Haishan Chen
Summary: A new integral method is developed to directly partition a horizontal velocity field without computing stream-function and velocity potential. By skipping intermediate steps, the new method eliminates finite-difference discretization errors and retains high accuracy for complex flow fields induced by severe weather and complex terrains. Numerical experiments demonstrate the superior performance and improved accuracy of the new integral method compared to previous methods. Thus, the new integral method is recommended unless stream-function and velocity potential need to be computed and used.
ATMOSPHERIC RESEARCH
(2023)
Article
Meteorology & Atmospheric Sciences
Xinguan Du, Haishan Chen, Qingqing Li, Xuyang Ge
Summary: Coastal urban areas play an important role in influencing the precipitation of landfalling tropical cyclones. High-resolution ensemble simulations of Typhoon Rumbia were conducted, showing that the inner-core rainfall is strengthened by approximately 10% due to the urban impact near the landfall. Further analyses indicate that low-level upward motion is crucial for precipitation evolution, and frictionally induced upward motion plays a decisive role in enhancing rainfall when the urban impacts are included.
ADVANCES IN ATMOSPHERIC SCIENCES
(2023)
Article
Meteorology & Atmospheric Sciences
Zhiyi Zhou, Juan Li, Haishan Chen, Zhiwei Zhu
Summary: Extreme high temperatures often occur in southwestern China, affecting the local ecology and economy. The accurate prediction of these extreme high-temperature days (EHDs) remains a challenge. This study identifies the spatiotemporal characteristics of EHDs and defines a domain-averaged EHD index for southwestern China. The study also finds meaningful precursors for EHDs and develops an empirical model for skillful prediction, achieving a correlation coefficient of 0.76.
ADVANCES IN ATMOSPHERIC SCIENCES
(2023)
Article
Meteorology & Atmospheric Sciences
Xuan Zhou, Jie Cao, Haishan Chen, Jisong Sun, Wei Zhao, Xiaobin Qiu, Linna Zhang, Hao Jing
Summary: The variability of regional persistent extreme precipitation events is increasing globally and in many regions. In 2021, the North China Plain experienced extremely strong and frequent but unevenly distributed extreme precipitation events. These events are influenced by multi-scale systems and sea-air interactions. The southern subdomain of the North China Plain exhibited significantly extreme precipitation characteristics in 2021.
ATMOSPHERIC RESEARCH
(2023)
Article
Meteorology & Atmospheric Sciences
Yidi Song, Haishan Chen
Summary: This study reveals the mechanisms underlying the interannual variability of spring land surface temperature over western Eurasia. The positive phase of North Atlantic tripole sea surface temperature anomalies in February can affect the circulation patterns and lead to local surface warming over western Eurasia in spring.
JOURNAL OF CLIMATE
(2023)
Article
Meteorology & Atmospheric Sciences
Yinshuo Dong, Haishan Chen, Xuan Dong
Summary: In the summer of 2020, a super Meiyu event occurred in the Yangtze River basin, and this study investigates the possible contributions of land surface processes to this extreme event. The study shows that antecedent soil moisture anomalies over the Indo-China Peninsula may have had a vital influence on the super Meiyu in 2020.
JOURNAL OF METEOROLOGICAL RESEARCH
(2023)
Article
Meteorology & Atmospheric Sciences
Yajing Qi, Haishan Chen, Siguang Zhu
Summary: Land-atmosphere coupling has a significant impact on low temperature extremes during fall and winter over southern Eurasia. It explains around 70% of temperature variability, increases near-surface air temperature, and reduces the frequency of LTEs. The coupling also alters atmospheric circulation, affecting the frequency and intensity of LTEs.
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2023)
Article
Meteorology & Atmospheric Sciences
Ning Xin, Botao Zhou, Haishan Chen, Shanlei Sun
Summary: A dipole pattern of spring vegetation in mid-high latitude Asia was detected based on the NDVI and empirical orthogonal function analysis. This pattern was found to be related to the autumn sea ice concentration anomalies in the Barents-Laptev Seas, with increased autumn SIC leading to increased spring NDVI over the west of Lake Baikal and decreased spring NDVI in the Russian Far East. Analyses further revealed the important role of SST anomalies in the Northeast Atlantic and the Northwest Pacific in this relationship.
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2023)
Article
Meteorology & Atmospheric Sciences
Shuyu Liu, Wenjian Hua, Liming Zhou, Haishan Chen, Miao Yu, Xing Li, Yazhu Cui
Summary: The biophysical effects of forest cover changes are frequently overlooked in climate policies, and there is a wide range of biophysical impacts of deforestation simulation results in state-of-the-art climate models. In this study, we used CMIP6-LUMIP simulations to examine the biophysical impacts of deforestation on global temperature and attributing the surface temperature change caused by deforestation to different biophysical effects at regional scales. The results show that models agree on the temperature responses to different biophysical factors in the tropics but diverge significantly in extratropical regions.
JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES
(2023)
Editorial Material
Multidisciplinary Sciences
Zhicong Yin, Botao Zhou, Mingkeng Duan, Haishan Chen, Huijun Wang
Article
Automation & Control Systems
Zhiguo Tan, Haishan Chen
Summary: This work generalizes the gradient-based design method to establish a finite-/fixed-time GNN for obtaining an online solution of the general linear matrix equation. Two nonlinear activation functions are introduced and utilized for constructing the GNN model, and their impacts on the convergence of both the GNN and the corresponding ZNN models are compared and analyzed. Theoretical analysis demonstrates that the GNN model and the ZNN model exhibit the same convergence property under the same activation function, and a tighter estimate of the upper bound of fixed convergence time is derived for the ZNN model. Comparative verification confirms the validity of the theoretical analysis.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(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.
