IEEE Transactions on Emerging Topics in Computational Intelligence最新文献_第8页

Dual-Population Evolution-Based Collaborative Optimization for Coordinated UAVs Mission Patrol Under Mountainous Environments 基于双种群进化的山地无人机任务巡逻协同优化

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-01-01 DOI: 10.1109/TETCI.2025.3645567

Yunhe Wang;Wenyuan Xiao;Dong Wang;Yajuan Zhang;Qingda Chen;Shengxiang Yang

{"title":"Dual-Population Evolution-Based Collaborative Optimization for Coordinated UAVs Mission Patrol Under Mountainous Environments","authors":"Yunhe Wang;Wenyuan Xiao;Dong Wang;Yajuan Zhang;Qingda Chen;Shengxiang Yang","doi":"10.1109/TETCI.2025.3645567","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3645567","url":null,"abstract":"Uncrewed aerial vehicles (UAVs) play a crucial role in visiting task points to support sector surveillance and maintain environmental and operational stability in mountainous mission patrols. Completing UAV patrol missions requires determining an optimal sequence for visiting task points and planning feasible paths to ensure the UAV safety in the presence of diverse peaks, valleys, and infeasible regions in mountainous environments. To address these issues, we propose a dual-population evolution-based collaborative optimization algorithm (DECOA) for coordinated UAV mission patrol, encompassing both UAV task assignment and path planning, in mountainous environments. First, we design a comprehensive model to represent mountainous environments accurately. Then, DECOA is proposed to address the UAV mission patrol optimization problem using a population-evolution framework applied to the constructed model. In DECOA, two distinct particle encoding schemes are introduced to efficiently convert task assignment and path planning solutions in UAV mission patrol models into representations suitable for optimization processing. Additionally, a dynamic dual-population search strategy is devised to balance exploration and exploitation capabilities. To enhance the global search capability of DECOA in promising regions, an adaptive collaboration mechanism is proposed. Furthermore, to evaluate the performance of DECOA, various scenarios are designed to conduct extensive experiments, focusing on UAV task assignment, path planning, and coordinated mission patrol. The experimental results demonstrate the superiority of the proposed algorithm and validate the effectiveness of the designed test scenarios.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1399-1410"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148513691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Intelligent Control of Semi-Active Friction Dampers for Seismic Control of Buildings: A Deep Reinforcement Learning Approach 用于建筑物地震控制的半主动摩擦阻尼器智能控制：一种深度强化学习方法

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-02-02 DOI: 10.1109/TETCI.2026.3656084

Mert Can Kurucu;Ercan Atam;Masoum M. Gharagoz;Mohamed Noureldin;Ibrahim Eksin;Müjde Güzelkaya

{"title":"Intelligent Control of Semi-Active Friction Dampers for Seismic Control of Buildings: A Deep Reinforcement Learning Approach","authors":"Mert Can Kurucu;Ercan Atam;Masoum M. Gharagoz;Mohamed Noureldin;Ibrahim Eksin;Müjde Güzelkaya","doi":"10.1109/TETCI.2026.3656084","DOIUrl":"https://doi.org/10.1109/TETCI.2026.3656084","url":null,"abstract":"Semi-active friction dampers (SAFDs), which can adaptively adjust the friction force, function as flexible semi-active control systems for the seismic protection of structures, including buildings exposed to strong ground accelerations. The effectiveness of SAFDs in dissipating seismic energy is highly dependent on the control strategy used to adjust the friction force of dampers. Due to the highly nonlinear and complex nature of buildings equipped with SAFDs, achieving optimal control is challenging and requires advanced adaptive control techniques. In this paper, we propose an intelligent control method based on deep reinforcement learning for the seismic control of buildings with SAFDs. We examine the potential of SAFDs to improve the seismic response of multi-story buildings and present a numerical case study to demonstrate the effectiveness of the proposed intelligent method. In this case study, the performance of the proposed intelligent control method is compared with two optimization methods and a reference controller: (i) friction dampers (FDs) using a constant friction force that is calibrated using a heuristic approach, (ii) FDs whose friction force is optimized through a Genetic Algorithm (GA), and (iii) SAFDs employing a rule-based approach for adaptive friction force adjustment. It was observed that the proposed DRL approach outperformed the alternatives, for example, achieving up to 63.7 % improvement over the rule-based controller. The core contribution of this study is the formulation of an MDP and reward-design framework that enables DRL to perform real-time friction control in SAFDs integrated into multi-story buildings.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1580-1593"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148504457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LogicICL: Towards Instructive Logical-Aware Demonstrations for Stimulating LLMs Mathematical Reasoning Ability 逻辑学：促进法学硕士数学推理能力的启发性逻辑意识演示

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-02-06 DOI: 10.1109/TETCI.2026.3654446

Yao He;Jiajun Liu;Wantao Wen;Yikai Guo;Hanchao Chen;Yue Ma;Wenjun Ke;Peng Wang

{"title":"LogicICL: Towards Instructive Logical-Aware Demonstrations for Stimulating LLMs Mathematical Reasoning Ability","authors":"Yao He;Jiajun Liu;Wantao Wen;Yikai Guo;Hanchao Chen;Yue Ma;Wenjun Ke;Peng Wang","doi":"10.1109/TETCI.2026.3654446","DOIUrl":"https://doi.org/10.1109/TETCI.2026.3654446","url":null,"abstract":"Large language models (LLMs) have demonstrated remarkable capabilities on various downstream tasks. However, mathematical problem-solving remains a formidable challenge due to its complex reasoningsteps. To address these issues, in-context learning (ICL) has emerged as a training-free paradigm that enhances LLMs reasoning ability by retrieving demonstrations. Existing ICL methods can be broadly categorized into similarity-based retrieval (SSR) and structural similarity-based retrieval (STR) methods. SSR methods retrieve demonstrations based on the similarity of texts in the semantic space, yet they fail to identify the underlying mathematical reasoning structure. To overcome this limitation, STR methods retrieve examples by aligning resoning structures between the target and candidate problems. However, they still face two major challenges: (1) misalignment in ensuring consistent reasoning paths between retrieved demonstrations and the target problem, and (2) inability to identify the underlying theorems required to solve the target problem. To alleviate these challenges, we propose LogicICL, a logic-aware ICL framework that retrieves examples with consistent inference paths and formulaic alignment by explicitly modeling symbolic reasoning. Specifically, (1) to address the misalignment in reasoning steps, we translate all problems into first-order logic (FOL) representations and mine circuit-level units for structured comparison of inference processes. (2) to capture the underlying theorems, we build logic circuit graphs from these units and apply graph-based similarity to retrieve demonstrations with consistent symbolic derivation and formulaic alignment. Experimental results show that LogicICL outperforms existing state-of-the-art baselines on the seven benchmarks, achieving average accuracy gains of 2.48%, 1.30%, 1.66% and 3.88% on the four backbones, respectively.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1669-1683"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508113","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Morphology Recognition Network via Contrastive Distillation for Detecting Rail Surface Defects 基于对比蒸馏的形态识别网络检测钢轨表面缺陷

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2025-08-07 DOI: 10.1109/TETCI.2025.3593905

Wujie Zhou;Yue Wu;Jiajun Meng;Xiena Dong

{"title":"Morphology Recognition Network via Contrastive Distillation for Detecting Rail Surface Defects","authors":"Wujie Zhou;Yue Wu;Jiajun Meng;Xiena Dong","doi":"10.1109/TETCI.2025.3593905","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3593905","url":null,"abstract":"Computer vision-based methods for rail defect detection can automatically identify defects on railroad surfaces to ensure transportation safety. However, these methods face challenges, such as defect shape irregularity and foreground-background confusion. In addition, rail transportation requires real-time detection to ensure safety, but existing methods are are challenging to implement on mobile devices. To address these difficulties, this study proposes a progressive morphology recognition network (MRNet) based on contrasting knowledge distillation (KD). First, we combine filter detection with neural networks to design a spatial morphology extraction module that carefully detects defect patterns in pixel space and directs the network to focus on salient targets in complex environments. Then, we use pixel domain-guided fusion to complete layer-by-layer refinement. Finally, we employ a bimodal input teacher network to distill a unimodal input student network. Additionally, we introduce a comparative learning approach that allows students to replicate the pixel-consistent learning exhibited by the teacher network. Experiments demonstrated that the proposed MRNet-KD significantly reduces computational costs while achieving satisfactory performance with only 16 million parameters. Additionally, MRNet-KD outperformed 14 state-of-the-art methods on the NEU RSDDS-AUG dataset, evaluated using six commonly used metrics.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1794-1804"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bilevel Multi-Objective Pricing Scheming in Partial-Offloading Wireless-Powered Cooperative MEC 部分卸载无线供电协同MEC的双层多目标定价策略

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2025-10-31 DOI: 10.1109/TETCI.2025.3623283

Weizhong Wang;Hai-Lin Liu;Kay Chen Tan

{"title":"Bilevel Multi-Objective Pricing Scheming in Partial-Offloading Wireless-Powered Cooperative MEC","authors":"Weizhong Wang;Hai-Lin Liu;Kay Chen Tan","doi":"10.1109/TETCI.2025.3623283","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3623283","url":null,"abstract":"Traditional pricing schemes of a wireless-powered cooperative mobile edge computing (WP-CoMEC) system assume that, local devices are bound to the service provider (SP), which not only limits the flexibility of customers to replace the SP in the future, but also raises concerns for potential customers who already have high-performance devices. Therefore, we present a novel pricing scheme for WP-CoMEC systems, allowing customers utilize their private devices as local computing resources. We focus on a scenario with extremely high computational complexity and low latency requirements for tasks and adopt partial offloading mode to tackle the challenge. We formulate the pricing scheme as a bilevel multi-objective optimization problem (BLMOP), where the upper-level decision-maker (the device owner) aims to minimize costs and the allocated computing capacity on local devices, while the lower-level decision-maker (the service provider) seeks to minimize costs and maximize profits. To address the computational challenges posed by this bilevel optimization, we propose a Pareto front prediction-based bilevel evolutionary multi-objective optimization algorithm (PFP-BLEMO). This algorithm employs a conditional generative adversarial network (cGAN) to predict the lower-level Pareto front, allowing for a more efficient search for optimal solutions. Our experimental results demonstrate the effectiveness and efficiency of the proposed approach, showing significant improvements in cost minimization and user experience in WP-CoMEC systems.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1137-1156"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148509659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

IEEE Transactions on Emerging Topics in Computational Intelligence Publication Information IEEE计算智能新兴主题汇刊

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-03-26 DOI: 10.1109/TETCI.2026.3671947

引用次数: 0

Unsupervised Domain Adaption for Multi-Modal Remote Sensing Image Semantic Segmentation via Asymmetric Boosting Fusion Network 基于非对称增强融合网络的无监督域自适应多模态遥感图像语义分割

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-01-12 DOI: 10.1109/TETCI.2025.3647688

Aihua Zheng;Jinchao Wang;Zi Wang;Chenglong Li;Bin Luo;Amir Hussain

{"title":"Unsupervised Domain Adaption for Multi-Modal Remote Sensing Image Semantic Segmentation via Asymmetric Boosting Fusion Network","authors":"Aihua Zheng;Jinchao Wang;Zi Wang;Chenglong Li;Bin Luo;Amir Hussain","doi":"10.1109/TETCI.2025.3647688","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3647688","url":null,"abstract":"Digital surface model (DSM) can assist optical images in semantic segmentation with complementary information and exhibit small domain shift between source and target domains for domain adaptation of remote sensing images. However, existing methods neglect the respective superiority of optical and DSM images, and diverse domain-independent features which play a critical role in domain adaptation for remote sensing image semantic segmentation. We propose a novel Asymmetric Boosting Fusion Network with a multi-head divergence discriminator, called ABFNet, which pursues the full utilization of both complementary benefits from DSM and optical images while capturing diverse domain-independent features. In particular, to capture effective class-discrimination features, we design an asymmetric mutual booster to enhance each modality according to the properties of another modality. It consists of a DSM-induced geometry booster to enhance the object position and boundary of the optical features, and an optical-induced content booster to enhance the semantic details of the DSM features. Then, we propose a multi-modal non-local fusion module to fuse the boosted features of both optical and DSM modalities. Moreover, to capture diverse domain-independent features in unsupervised domain adaptation, we design a multi-head divergence discriminator with two domain classification heads, and a parameter divergence loss constraining the inconsistency between the classifiers.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1482-1497"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148511011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Learning Robot Dynamic Cloth Manipulation Based on Differentiable Dynamics Optimization and Motion Transfer 基于可微动力学优化和运动传递的机器人动态布艺学习

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2025-11-10 DOI: 10.1109/TETCI.2025.3623331

Hengyi Sim;Junjia Liu;Donatien Delehelle;Xingyu Wu;Darwin G. Caldwell;Fei Chen

{"title":"Learning Robot Dynamic Cloth Manipulation Based on Differentiable Dynamics Optimization and Motion Transfer","authors":"Hengyi Sim;Junjia Liu;Donatien Delehelle;Xingyu Wu;Darwin G. Caldwell;Fei Chen","doi":"10.1109/TETCI.2025.3623331","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3623331","url":null,"abstract":"Dynamic manipulation of soft objects are often employed by humans in everyday life. In robotics, cloth manipulation serves as a commonly used test case for researching the dynamic manipulation of soft objects by robots, thereby improving the robot’s task efficiency with extended control over uncontacted regions of the cloth. However, dynamic manipulation of cloths introduces challenges such as controlling the high dimensional state space of cloth resulting in complex dynamics, and the inherent difficulty of planning the non-prehensile manipulation. In this work, we propose a cloth dynamic manipulation policy learned via an adversarial imitation learning pipeline based on realistic cloth dynamics trained from real cloth deformation ground truth through the real2sim dynamics exploration module. The real2sim module leverages gradient based optimization to learn a realistic cloth model that performs well under significant deformations, while generalizing across different materials. Imitation learning is then used to learn manipulation styles directly from observing realistic cloth states generated from just a single expert manipulation path. Several experiments are then designed to evaluate the generalization ability of the proposed method, and have been deployed on a humanoid robot to generate bimanual dynamic manipulation motions for cloth handling.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1641-1653"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148511181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring the Potential of Class-Specific Encoding for Few-Shot Object Detection 探索类特定编码的潜力，为少数镜头目标检测

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2026-01-05 DOI: 10.1109/TETCI.2025.3647629

Wei Zhou;Zongling Li;Shu Hu;Shanmin Yang;Tao Wu;Siwei Lyu;Ying Fu;Xi Wu;Xin Wang

{"title":"Exploring the Potential of Class-Specific Encoding for Few-Shot Object Detection","authors":"Wei Zhou;Zongling Li;Shu Hu;Shanmin Yang;Tao Wu;Siwei Lyu;Ying Fu;Xi Wu;Xin Wang","doi":"10.1109/TETCI.2025.3647629","DOIUrl":"https://doi.org/10.1109/TETCI.2025.3647629","url":null,"abstract":"Recently, the object detection methods based on few shot learning have significantly advanced the field of object detection, because the methods are easy to achieve excellent detection results when there are few training samples. However, detecting novel classes with existing methods often requires time-consuming model retraining. Moreover, the detection performance of new models for base classes may decrease. In order to solve the above problems, this paper proposes a new few-shot object detection model, which can efficiently detect the novel classes introduced without fine-tuning the model. This model enables simultaneous detection of both novel and base classes, effectively mitigating catastrophic forgetting. The model incorporates a base class detector, augmented with an additional contrastive branch to extract class representation information. This decouples object localization and classification, leading to a marked improvement in the model’s generalization performance for novel classes. In addition, we investigate the effectiveness of self-supervised and supervised contrastive losses for class-specific encodings in our framework.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"10 2","pages":"1441-1453"},"PeriodicalIF":6.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148511356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Deep Hamming Hashing Integrating Transformer and Attention Mechanism 深度汉明哈希集成变压器和注意机制

IF 6 3区计算机科学

IEEE Transactions on Emerging Topics in Computational Intelligence Pub Date : 2026-04-01 Epub Date: 2025-12-08 DOI: 10.1109/TETCI.2025.3630464

Xinyi Wang;Dongsheng Guo;Weidong Zhang;Zehua Jia;Shan Xue;Wenbo Zhang

引用次数: 0