Article
Computer Science, Artificial Intelligence
Jun-qing Li, Zheng-min Liu, Chengdong Li, Zhi-xin Zheng
Summary: This article proposes an improved artificial immune system (IAIS) algorithm to solve a special case of the flexible job shop scheduling problem (FJSP), where the processing time of each job is a nonsymmetric triangular interval T2FS (IT2FS) value. The algorithm shows enhanced abilities in handling high levels of uncertainty and asymmetric triangular interval values. Through novel affinity calculation methods, problem-specific initialization heuristics, local search approaches, and population diversity heuristics, the algorithm demonstrates improved exploitation and exploration capabilities.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Dogan Corus, Pietro S. Oliveto, Donya Yazdani
Summary: This study proposes modifications to traditional hypermutation operators to improve efficiency during the exploitation phase while maintaining effective explorative characteristics. By using a stochastic parabolic distribution for fitness function sampling, wasted function evaluations are reduced. The proposed operators are rigorously proven to be effective for benchmark functions and show linear speed-ups in identifying high-quality approximate solutions to NP-Hard problems. Through comparative performance studies, it is concluded that a power-law distribution for the parabolic evaluation scheme is the best compromise in scenarios with limited problem knowledge.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Environmental Sciences
Li-Feng Huang, Cheng-Guo Liu, Zhi-Peng Wu, Li-Jun Zhang, Hong-Guang Wang, Qing-Lin Zhu, Jie Han, Ming-Chen Sun
Summary: Using intelligent optimization algorithms, this study developed a new passive remote sensing technology for atmospheric ducts by monitoring automatic identification system (AIS) signals at sea. The results showed that the particle swarm optimization (PSO) algorithm had the best inversion performance. Further statistical analysis of the inversion results under different parameters confirmed the effectiveness of the proposed algorithm. However, the improvement trend gradually slowed, indicating the need to balance inversion time consumption and inversion effect in practical applications.
Article
Engineering, Environmental
Yuman Yao, Yiyang Dai, Jinsong Zhao
Summary: An enhanced dynamic artificial immune system based on simulated vaccine and correlation coefficient methods (SV-CCDAIS) is proposed to improve the process safety of chemical systems with extreme absence of data. By using simulated vaccine, dynamic time warping, and different dynamic correlation measurement methods, the proposed method achieves higher fault diagnosis accuracy and shorter diagnosis time compared to the comparative method.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2022)
Review
Immunology
Coralie Backlund, Sasan Jalili-Firoozinezhad, Byungji Kim, Darrell J. Irvine
Summary: Modulation of the immune system is crucial in therapeutic strategies and vaccine development. Biomaterials are being developed to accurately deliver stimulatory cues to specific cells at the right time and place, ensuring safe and effective immunotherapy. This article discusses the uses of biomaterials in enhancing immune modulation and presents evidence from clinical and preclinical studies on how biomaterials can impact the development of new therapeutics. The focus is on understanding immunological mechanisms and in vivo immune system modulation, as well as overcoming challenges in translating these technologies to clinical practice.
ANNUAL REVIEW OF IMMUNOLOGY
(2023)
Article
Thermodynamics
Siyuan Yang, Junqi Yu, Zhikun Gao, Anjun Zhao
Summary: The air-conditioning water system deviates from its optimum state under partial load, so it is crucial to dynamically adjust the operating parameters of the equipments in the system to maximize energy efficiency. An improved parallel artificial immune system (IPAIS) algorithm is proposed to determine the optimal parameters under different loads. The algorithm is tested using a simulation experiment on an actual air-conditioning water system, and it achieves superior optimization results compared to other algorithms in terms of convergence, robustness, and computational complexity.
ENERGY CONVERSION AND MANAGEMENT
(2023)
Article
Economics
Dan Liu, Pengyu Yan, Ziyuan Pu, Yinhai Wang, Evangelos Kaisar
Summary: Research has established a model and optimized E-grocery delivery, which can achieve balance among economic costs, environmental effects, and customer satisfaction.
TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
(2021)
Article
Computer Science, Artificial Intelligence
Liang Xi, Rui-Dong Wang, Zhi-Yu Yao, Feng-Bin Zhang
Summary: The artificial immune system (AIS) is an important branch of artificial intelligence technology, and its application effects rely on the generation, evolution, and detection of detectors. The current real-valued detectors face issues like slow generation speed, holes in nonself region, overlapping redundancy, etc., requiring better adaptive modeling for solutions.
IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Bahareh Etaati, Zahra Ghorrati, Mohammad Mehdi Ebadzadeh
Summary: In this paper, a novel cooperative coevolutionary memory-based artificial immune system is proposed for dynamic optimization problems. The algorithm decomposes the population into subpopulations and utilizes short-term memory and long-term memory to evaluate and store essential information. The proposed approach demonstrates competitive performance in optimizing dynamic problems.
APPLIED SOFT COMPUTING
(2022)
Article
Computer Science, Artificial Intelligence
Lingjie Li, Qiuzhen Lin, Zhong Ming
Summary: This paper presents a comprehensive survey on the multi-objective immune algorithm (MOIA), which is a heuristic algorithm based on the artificial immune system model. MOIAs can be classified into three main categories: multi-objective optimization problems (MOPs), dynamic MOPs, and constrained MOPs. The characteristics, principles, theoretical analyses, and performance of MOIAs in solving various types of MOPs are discussed. The paper concludes with a brief summary of the current drawbacks, challenges, and future directions for MOIAs.
Article
Chemistry, Multidisciplinary
Haochuan Wan, Junyi Zhao, Li-Wei Lo, Yunqi Cao, Nelson Sepulveda, Chuan Wang
Summary: The sensory-memory system is crucial in the formation of human intelligence, enabling perception, interaction, and evolution with the environment. The electronic implementation of this system is driving the development of environment-interactive artificial intelligence, potentially expanding the application of AI in human-computer interaction. The biomimetic intelligence demonstrated in this system shows promise in advancing multimodal, user-environment interactive AI.
Article
Computer Science, Information Systems
Shijing Ma, Yunhe Wang, Shouwei Zhang
Summary: The authors of this study propose a crowding artificial bee colony (IABC) algorithm that combines the concepts of crowding and search solution exploration for solving multimodal functional optimization problems. Experimental results demonstrate that the method is effective and efficient.
JOURNAL OF ORGANIZATIONAL AND END USER COMPUTING
(2022)
Review
Immunology
Andi Zhang, Tianyuan Zou, Dongye Guo, Quan Wang, Yilin Shen, Haixia Hu, Bin Ye, Mingliang Xiang
Summary: Noise can have significant impacts on the immune system, leading to a variety of physical and mental disorders. The duration and intensity of noise exposure can determine the effects on the immune system, with short-term or low-intensity noise enhancing immune function, while long-term or high-intensity noise suppressing it.
FRONTIERS IN IMMUNOLOGY
(2021)
Article
Automation & Control Systems
Mu-Yen Chen, Hsin-Te Wu
Summary: As intelligent transportation systems evolve, this article focuses on the fundamental components of vessels, harbors, and ship-shore information-communication technology applications. By integrating artificial intelligence and the 5G network, the research aims to ensure ship safety through information sharing and the implementation of geofencing technology. The proposed security mechanism detects malicious attacks and improves vessel safety through the use of the 5G network and sensors. The performance analysis and experimental results demonstrate the effectiveness of the network security approach and geofencing technology introduced in this article.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Engineering, Multidisciplinary
Petar Cisar, Sanja Maravic Cisar, Brankica Popovic, Kristijan Kuk, Igor Vukovic
Summary: This paper discusses the application of artificial immune networks in continuous function optimizations, and analyzes the performance of immunological algorithms. It was found that the CLIGA algorithm has the fastest convergence and best score, while the opt-IA algorithm achieved the lowest total number of iterations within the defined run time.
ACTA POLYTECHNICA HUNGARICA
(2022)
Article
Automation & Control Systems
Zhichao Sun, Hang Ren, Huarui Sun, Gary G. Yen, Junjie Wu, Jianyu Yang
Summary: This article investigates the terminal trajectory planning for synthetic aperture radar (SAR) imaging guidance. A chronological iterative search framework (CISF) is proposed to solve the trajectory planning problem by decomposing it into subproblems and utilizing the optimization results of preceding subproblems. Experimental studies show the effectiveness and superiority of CISF compared to other methods.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Automation & Control Systems
Zhichao Sun, Gary G. Yen, Junjie Wu, Hang Ren, Hongyang An, Jianyu Yang
Summary: This article proposes a mission planning framework for an energy-efficient passive UAV radar imaging system and introduces a path planning method called Sub-DiCoS. The method adjusts the UAV's flight path and utilizes differential evolution and the whole-stage best guidance technique to achieve optimized imaging and communication performance in an energy-efficient manner.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Automation & Control Systems
Weifeng Gao, Zhifang Wei, Maoguo Gong, Gary G. Yen
Summary: This article proposes a decomposition differential evolution algorithm based on radial basis function to solve multimodal optimization problems. The algorithm decomposes the problem into multiple global optimization subproblems and solves them using population update strategy and local RBF surrogate models.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Automation & Control Systems
Zhichao Sun, Hang Ren, Gary G. Yen, Tianfu Chen, Junjie Wu, Hongyang An, Jianyu Yang
Summary: In this article, an evolutionary algorithm with constraint relaxation strategy based on differential evolution algorithm (CRS-DE) is proposed to solve Highly Constrained Multiobjective Optimization Problems (HCMOPs). The algorithm relaxes the constraints by dividing the infeasible solutions into semifeasible subpopulation (SF) and infeasible subpopulation (IF), and devises corresponding reproduction and selection strategies for SF, IF, and feasible subpopulations. To prevent premature convergence, a mobility restriction mechanism is developed to restrict the individuals in SF and IF from entering the feasible subpopulation and enhance the diversity of the whole population.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Fei Zou, Qingxin Guo, Gary G. Yen
Summary: In this study, a dynamic multiobjective optimization model is proposed to address an operations decision problem in the steel industry. A multiobjective evolutionary algorithm called KPLSEA is developed to solve the problem, which significantly reduces computational cost and promotes quick convergence. Extensive experiments validate the effectiveness and practicality of the proposed algorithm in dynamic environments.
APPLIED SOFT COMPUTING
(2023)
Article
Engineering, Electrical & Electronic
Zhichao Sun, Junjie Wu, Gary G. Yen, Zheng Lu, Jianyu Yang
Summary: This paper investigates the performance and implementation of an energy-efficient passive UAV radar imaging system. Equipped with a synthetic aperture radar (SAR) receiver, the system passively reuses the backscattered signal of an external illuminator, achieving SAR imaging and data communication. The article presents the system concept, analyzes the imaging performance and feasibility for typical illuminators, and establishes a set of mission performance evaluators for comprehensive assessment.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Hong Li, Weifeng Gao, Jin Xie, Gary G. Yen
Summary: This article presents a method for automatically designing the network architecture of multilayer perceptron (MLP) neural networks and optimizing network parameters using a multiobjective bilevel programming model. The upper level constructs a multiobjective optimization problem to obtain a set of Pareto optimal network structures for the MLPs, while the lower level solves a single-objective optimization problem to search for the optimum network parameters. A novel multiobjective hierarchical learning algorithm (MOHLA) is proposed to efficiently deal with this model, and a selective ensemble strategy is adopted to improve identification accuracy. Experimental results confirm the excellent performance of MOHLA.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Siyi Li, Yanan Sun, Gary G. Yen, Mengjie Zhang
Summary: With the rise of smart electronics and mobile devices, existing high-accuracy CNN models are difficult to apply due to limited resources. In this article, we propose an automatic method for designing CNN architectures under constraint handling, which effectively searches for optimal network models meeting preset constraints.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Zequn Sun, Wei Hu, Chengming Wang, Yuxin Wang, Yuzhong Qu
Summary: Entity alignment, the discovery of identical entities across different knowledge graphs, is a critical task in data fusion. Existing entity alignment methods lack robustness to long-tail entities and the absence of entity names or relation triples. This paper proposes a robust and adaptive entity alignment method that does not require relations, attributes, or names, achieving state-of-the-art performance even in challenging settings without relations and names.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Information Systems
Zaixing Sun, Boyu Zhang, Chonglin Gu, Ruitao Xie, Bin Qian, Hejiao Huang
Summary: In this article, a hybrid heuristic algorithm called ET2FA is proposed to solve deadline-constrained workflow scheduling in the cloud. With new features such as hibernation and per-second billing, ET2FA can generate efficient and economical scheduling schemes. Extensive simulation experiments show that ET2FA outperforms state-of-the-art algorithms.
IEEE TRANSACTIONS ON SERVICES COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Qiyu Sun, Gary G. G. Yen, Yang Tang, Chaoqiang Zhao
Summary: Monocular depth estimation, achieved through deep learning, is a fundamental task in environmental perception. However, trained models often exhibit degraded performance when applied to new datasets due to dataset differences. We propose a meta-learning framework with an adversarial depth estimation task to improve the transferability and alleviate meta-overfitting issues of self-supervised monocular depth estimation models. Our method demonstrates fast adaptation to new domains and achieves comparable results to state-of-the-art methods after only 0.5 epoch of training.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Junhao Huang, Bing Xue, Yanan Sun, Mengjie Zhang, Gary G. Yen
Summary: Neural architecture search (NAS) is a popular research topic in deep learning community due to its potential in automating the construction of deep models. Among various NAS approaches, evolutionary computation (EC) stands out for its capability of gradient-free search. However, most current EC-based NAS approaches have the limitation of discrete evolution, making it difficult to handle the number of filters for each layer flexibly. Additionally, EC-based NAS methods are criticized for their inefficiency in performance evaluation, often requiring full training of hundreds of candidate architectures. This work proposes a split-level particle swarm optimization (PSO) approach to address these issues and achieves superior performance on image classification benchmarks.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Haokai Hong, Min Jiang, Gary G. Yen
Summary: The large-scale multiobjective optimization problem (LSMOP) involves optimizing multiple conflicting objectives and hundreds of decision variables. Existing algorithms often focus on improving performance but pay little attention to improving insensitivity. We propose an evolutionary algorithm based on Monte Carlo tree search to improve the performance and insensitivity of solving LSMOPs.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Zehua Sun, Qiuhong Ke, Hossein Rahmani, Mohammed Bennamoun, Gang Wang, Jun Liu
Summary: Human Action Recognition (HAR) aims to understand human behavior and assign labels to actions. Various data modalities, such as RGB, skeleton, depth, infrared, point cloud, event stream, audio, acceleration, radar, and WiFi signal, can be used to represent human actions. Many studies have investigated different approaches for HAR using these modalities. This article presents a comprehensive survey of recent progress in deep learning methods for HAR based on input data modality, covering single and multiple modalities and fusion-based and co-learning-based frameworks.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Xuewen Bing, Wenqi Ren, Yang Tang, Gary G. Yen, Qiyu Sun
Summary: This study presents an in-air to underwater image enhancement framework that overcomes the limitations of underwater synthetic datasets by utilizing color correction and domain adaptation steps, resulting in a more natural appearance and better generalization.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)