Applied Intelligence最新文献

{"title":"Optimizing time slot allocation in complex multi-dock truck loading and unloading operations using evolutionary algorithms","authors":"Enol García González,&nbsp;José R. Villar,&nbsp;Javier Sedano,&nbsp;Camelia Chira","doi":"10.1007/s10489-026-07334-7","DOIUrl":"10.1007/s10489-026-07334-7","url":null,"abstract":"<div><p>This study focuses on an industrial facility’s time-slot allotment (TSA) problem, where loading and unloading docks are assigned to the incoming lorries, reducing the number of them waiting for service. Several constraints apply to the different dock stations, including disparate timetables and task duration, various capacities of simultaneous services, etc. Evolutionary algorithms cope with the enormous variability of the combinatorial problem, maximizing the number of accepted lorries while decreasing the queue at the entrance. Interestingly, the problem’s structure led to unconventional operator probabilities, which also analyzes the evolutionary techniques and operators included in this study. Comparative analysis with state-of-the-art methods highlights the algorithm’s effectiveness, though computational demands rise with population size.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-026-07334-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"LHPS-Net: Lipschitz hierarchical pyramid for medical image registration with spatial recurrent encoders","authors":"Chao Fan,&nbsp;Huijun Zhao,&nbsp;Xinru Zhu,&nbsp;Zhihui Xuan,&nbsp;Zhentong Zhu,&nbsp;Rangyang Li,&nbsp;Xiaodong Guo","doi":"10.1007/s10489-026-07344-5","DOIUrl":"10.1007/s10489-026-07344-5","url":null,"abstract":"<div><p>Medical image registration plays a crucial role in tumor growth monitoring, radiotherapy planning, and disease diagnosis. Recently, Transformer-based networks have been widely adopted in unsupervised image registration, achieving improved registration accuracy. However, the use of attention mechanisms significantly increases model complexity with substantial computational and memory costs, failing to meet the real-time requirements of medical image registration. To address these challenges, this paper proposes a Hierarchical Pyramid Network with Lipschitz Continuity Constraint and Spatial Recurrent Encoding Module for Medical Image Registration (LHPS-Net). The model employs a hierarchical pyramid architecture that enhances registration performance through Lipschitz continuity constraints and spatial recurrent encoding. Specifically, unlike traditional deep learning-based approaches, LHPS-Net independently extracts features from fixed and moving images and progressively generates deformation fields through hierarchical decoders, which facilitates large-deformation image registration. The spatial recurrent encoding module replaces conventional convolutional blocks, utilizing SLK (Spatial Locally-connected Kernel) to capture contextual information with lower computational cost while effectively focusing on multi-scale features. The LC-Def block incorporates Lipschitz continuity constraints to help achieve diffeomorphic deformation. To validate the performance of our proposed network, we conducted comprehensive comparisons with existing methods on public datasets including IXI, LPBA, and OASIS. Experimental results demonstrate that LHPS-Net achieves superior registration performance under common evaluation metrics.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DFB: AData-Free, Low-Budget, and High-EfficacyClean-Label Backdoor Attack","authors":"Binhao Ma,&nbsp;Jiahui Wang,&nbsp;Dejun Wang,&nbsp;Bo Meng","doi":"10.1007/s10489-026-07328-5","DOIUrl":"10.1007/s10489-026-07328-5","url":null,"abstract":"<div><p>With the continuous advancement of machine learning, backdoor attacks have become significant security concerns. These attacks embed malicious triggers into models, allowing them to function normally under standard conditions but causing them to exhibit malicious behavior when triggered. In backdoor attacks, accurately labeling injected data is essential for evading basic detection mechanisms. Clean-label backdoor attacks achieve this goal by preserving the correct labels while embedding imperceptible triggers. However, they typically require access to the victim’s training dataset, which is often unattainable due to the decentralized nature of data collection. This limitation significantly restricts the real-world applicability of these attacks. To overcome the limitations of the existing clean-label attacks, we propose DFB, which is a data-free, low-budget, and high-efficacy backdoor attack. DFB requires only target-class knowledge, eliminating the need for access to victim datasets. By leveraging POOD data to generate ’hard-to-learn’ target features, which the model effectively learns, we inject minimal samples to launch the attack. Experiments conducted on CIFAR-10, Tiny-ImageNet, and TSRD show that DFB achieves superior attack success rates with minimal poisoning rates (0.1%, 0.025%, and 0.5%), significantly outperforming LC, HTBA, BadNets, and Blend. Furthermore, our findings reveal that DFB poses a formidable challenge to four established backdoor defense algorithms, indicating its potential for use as a robust tool in advanced clean-label attack strategies.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Overcoming modality laziness in multi-modal multi-instance learning for immune repertoire classification","authors":"Yu-Xuan Zhang,&nbsp;Zhengchun Zhou,&nbsp;Hanjie Luo,&nbsp;Weisha Liu,&nbsp;Ming Li","doi":"10.1007/s10489-026-07295-x","DOIUrl":"10.1007/s10489-026-07295-x","url":null,"abstract":"<div><p>Immune repertoire classification (IRC) is a typical multi-instance learning (MIL) problem that is widely applied in fields such as vaccine development and drug discovery in computational biology. In existing MIL-IRC methods, the input modalities consist of amino acid (AA) sequences and V(D)J genes. However, common practices typically involve directly connecting the two modalities or treating them equally by merging them in the hidden space. This leads to the problem of modality laziness, where some modalities appear more dominant than others during multi-modal learning. To address this problem, this paper proposes a two-stage IRC method, called Overcoming Modality Laziness in Multi-modal Multi-instance Learning (OML-M3IL). In stage 1, we use alternating unimodal adaptation to learn from the two modalities separately, ensuring that the shared classifier works reliably with each modality alone. In stage 2, we also use a similarity loss to reduce the difference between modalities and fuse them. In addition, gradient orthogonalization, label disambiguation, and dynamic fusion prediction techniques are used to perform the IRC task more efficiently. Experiments were conducted on CMV and Cancer datasets, and the results verify that OML-M3IL yields consistently competitive or better performance than existing MIL-IRC methods while explicitly mitigating modality laziness.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving text classification on small-sized, imbalanced datasets with selective few-random-shot augmentation","authors":"Lingshu Hu,&nbsp;Peter Dolan,&nbsp;Can Li,&nbsp;Wenbo Wang,&nbsp;Bin Pang,&nbsp;Yi Shang","doi":"10.1007/s10489-026-07299-7","DOIUrl":"10.1007/s10489-026-07299-7","url":null,"abstract":"<div><p>Small and highly imbalanced text datasets often produce brittle classifiers because scarce minority examples provide limited coverage of class semantics, while skewed class distributions bias decision boundaries toward the majority class. We propose Selective Few-Random-Shot Augmentation (SFRSA), a generate-then-select framework for low-resource imbalanced text classification. SFRSA first over-generates minority-class candidates through few-random-shot LLM prompting, then selects a fixed-budget subset by optimizing the trade-off between relevance to minority examples and diversity among selected samples in embedding space. To operationalize this idea, we develop and compare three selection modules: centroid-guided clustering (CGC), determinantal point process (DPP) selection, and a training-aware influence-utility DPP (IU-DPP) variant. Across three benchmark datasets under training sizes of 500, 1000, and 2000 and imbalance ratios of 4:1 and 9:1, SFRSA consistently improves minority-sensitive performance over classic text augmentation methods and standard LLM-based generation baselines, with the largest gains appearing in the most data-scarce settings. A similarity–diversity analysis links the selection objective to downstream gains, showing that SFRSA preserves minority semantics while avoiding near-duplicate generations. The method is model-agnostic and can be integrated into existing NLP pipelines to improve classification in low-resource, imbalanced settings.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-026-07299-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A triple-head network with loss-aware label assignment for object detection","authors":"Wenjie Lin,&nbsp;Jun Chu,&nbsp;Lu Leng,&nbsp;Xingbo Dong","doi":"10.1007/s10489-026-07329-4","DOIUrl":"10.1007/s10489-026-07329-4","url":null,"abstract":"<div><p>Label assignment strategy plays a vital role in supervised object detection systems. However, the handcrafted label assignment used in most previous studies is usually suboptimal because the assignment strategy is not adaptive to the training objective. Considering the limitations of handcrafted label assignments, we propose a novel object detection framework consisting of a triple-head network and a joint-loss-aware label assignment strategy. The foregrounds and backgrounds are first determined based on a ground truth (GT) box. A label assignment strategy was proposed, which leveraged the joint loss of classification and regression as a unified sampling criterion to distinguish between positive and negative samples, while dynamically assigning optimal GT label for supervised learning. In addition, ambiguous samples from overlapping areas of multiple GT boxes were sampled based on a loss-weighting mechanism. We added a pre-regression branch to the head of FCOS, forming a triple-head detection network alongside the original classification and regression branches. The pre-regression branch performed coarse localization prediction on the object, provided additional contextual information for the classification and regression branches. Extensive experiments showed that the proposed method can effectively improve the detection performance for small and medium objects. And the proposed method achieved an average precision (AP) of 45.0% on the MS COCO test-Dev2017 dataset, achieving performance comparable to state-of-the-art methods.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148323122","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-agent graph reinforcement learning for real-time traffic signal control with spatio-temporal awareness","authors":"Tarek Amine Haddad,&nbsp;Abdessalam Hattab","doi":"10.1007/s10489-026-07336-5","DOIUrl":"10.1007/s10489-026-07336-5","url":null,"abstract":"<div><p>This paper presents a decentralized reinforcement learning framework for adaptive traffic signal control that integrates Graph Attention Networks (GAT) to model spatial dependencies in urban traffic networks. In the proposed approach, each intersection is controlled by an independent agent that observes local traffic conditions such as normalized queue lengths, elapsed phase durations, and aggregated neighboring congestion indicators and learns adaptive signal control policies under decentralized execution. Graph-based spatial aggregation enables localized information exchange among neighboring intersections, allowing agents to capture contextual traffic interactions without relying on centralized coordination. The resulting context-aware state representations are combined with temporal policy learning to improve decision quality in dynamically evolving traffic environments. The effectiveness of the proposed method is evaluated through extensive simulations on both synthetic grid networks and a real-world urban scenario from Batna, Algeria. Experimental results demonstrate consistent improvements in key traffic performance metrics, including average queue length, waiting time, throughput, and emissions, when compared to operational and decentralized learning baselines. Furthermore, the framework exhibits stable learning dynamics, faster convergence behavior, and strong scalability across different network topologies. These findings highlight the potential of graph-based decentralized reinforcement learning for efficient and adaptive urban traffic management.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Incomplete multi-view clustering based on discriminative anchor-feature mining-assisted adaptive sample completion","authors":"Mengying Xie,&nbsp;Gaojian Luo,&nbsp;Yi Xu,&nbsp;Wenjun Zhao,&nbsp;Xiaolan Liu","doi":"10.1007/s10489-026-07330-x","DOIUrl":"10.1007/s10489-026-07330-x","url":null,"abstract":"<div><p>Incomplete multi-view clustering (IMC) has gained increasing attention for the frequent occurrence of missing samples in real-world multi-view data. And adaptive sample completion has become a widely adopted strategy in handling IMC, as it facilitates the recovery of latent information from incomplete views. Unfortunately, most existing adaptive sample completion-based IMC methods still have the high computational cost. Although anchor-based strategies have been utilized to reduce computational complexity, existing methods typically minimize the anchor-based reconstruction error to iteratively complete the data matrix, thereby underutilizing latent discriminative structures and limiting the completion performance. To address both the high computational complexity and the underutilization of latent discriminative features in existing sample completion-based IMC methods, we propose a novel IMC method, DAM-ASC, based on discriminative anchor-feature mining-assisted adaptive sample completion. The proposed method integrates an adaptive dual reconstruction mechanism with a local geometric preservation strategy, enabling a mutual enhancement between sample completion and discriminative representation learning. Specifically, a dual reconstruction representation based on anchors and latent discriminative feature mining is designed to adaptively complete missing samples with improved efficiency and accuracy. Furthermore, a local manifold constraint based on anchor-sample-guided local geometric structure preservation is introduced to preserve the intrinsic data structure and enhance the reliability of the reconstructed samples. Experimental results on multiple public datasets demonstrate that DAM-ASC achieves superior clustering performance compared to several state-of-the-art IMC methods.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148281645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparative analysis of models capturing foot trajectory complexity across ambulation modes: application to robotic prostheses","authors":"Hamza Al Kouzbary,&nbsp;Mouaz Al Kouzbary,&nbsp;Jingjing Liu,&nbsp;Hamam Mokayed,&nbsp;Nooranida Arifin,&nbsp;Noor Azuan Abu Osman","doi":"10.1007/s10489-026-07315-w","DOIUrl":"10.1007/s10489-026-07315-w","url":null,"abstract":"<div><p>Accurate prediction of foot pattern throughout various terrains is essential for achieving stable and adaptive control in powered lower-limb prostheses. In this study, a range of predictive models, including regression-based approaches, ensemble machine learning methods, and recurrent neural networks (RNNs), were systematically compared to determine their capability in generating oscillatory gait signals from tibial angular position while walking on varied terrain. The training and testing data were collected from ten healthy individuals while level-ground walking, ascending and descending stairs. Statistical models had poor accuracy and generalization, while ensemble methods (Gradient Boosting, Histogram-Based Gradient Boosting, and eXtreme Gradient Boosting) had moderate performance but remained sensitive to outliers and inter-subject variation. In contrast, recurrent architectures, i.e., long short-term memory networks (LSTM), had the highest predictive accuracy, with a mean correlation of 0.88, and a low root mean square error of 0.09 rad. Although the LSTM-based method provided slightly lower accuracy than some of the control methods inspired by the central pattern generators in the literature, it required only prosthesis-embedded sensors, avoiding the need for sensors on the intact or residual limb. Moreover, the compact network size and low computational load make it well-suited for embedded deployment, reducing cost, power consumption, and user burden. These findings place RNN-based approaches in line with ongoing research trends toward dynamic pattern generator–inspired controllers, offering robust, volitional-like control while maintaining practicality for real-world implementation.</p><h3>Graphical Abstract</h3><div><figure><div><div><picture><source><img></source></picture><span>The alternative text for this image may have been generated using AI.</span></div></div></figure></div></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10489-026-07315-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148261819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MSTN-ISNet: A probability-guided spatio-temporal decoding for alleviating imbalance in sleep stage classification","authors":"Dongrui Gao,&nbsp;Shibing Li,&nbsp;Haokai Zhang,&nbsp;Qian Wu,&nbsp;Zongyao Peng,&nbsp;Shihong Liu,&nbsp;Liping Xie,&nbsp;Shaofei Ying,&nbsp;Jiaxin Xie","doi":"10.1007/s10489-026-07314-x","DOIUrl":"10.1007/s10489-026-07314-x","url":null,"abstract":"<div><p>Sleep staging is a critical process in the study and evaluation of sleep architecture, with its accuracy directly influencing the effectiveness of sleep disorder diagnosis. In temporal sleep staging, a common challenge arises from class imbalance—particularly, the significantly smaller number of N1 stage samples compared to other stages. Existing approaches have primarily focused on oversampling and expanding N1 stage data to address this issue. However, such methods may disrupt the temporal structure of the original signals and potentially lead to overfitting. To overcome these limitations, this paper proposes a Multi-Modal Spatio-Temporal Network for Imbalanced Sleep Staging (MSTN-ISNet), designed to perform data augmentation and enhancement for underrepresented sleep stages, thereby effectively alleviating the problem of insufficient training data for minority classes. First, we introduce a spatio-temporal statistics-driven time-frequency data augmentation module to enhance the representation of minority class samples. Next, we develop a deep interactive cross-modal feature encoding module to extract multi-modal spatio-temporal dynamic features from polysomnography (PSG) signals. Finally, a Markov chain entropy-weighted intelligent decision module is employed to calculate state transition probabilities among sleep stages, capturing the dynamic transitions between stages. Furthermore, to address the issue of class confusion, we implement strategies that reward minority classes, calculate inter-stage similarity, and utilize a weighted cross-entropy loss function to enhance the classification performance for minority stages. We evaluate the model on three benchmark datasets. Experimental results demonstrate classification accuracies of 90.8%, 90.4%, and 77.4%, respectively, outperforming existing state-of-the-art (SOTA) models. Furthermore, classification accuracies for N1-stage classification reached 85.4%, 92.5%, and 84.4%, respectively, exhibiting outstanding performance in minority class identification and surpassing all current advanced models.</p></div>","PeriodicalId":8041,"journal":{"name":"Applied Intelligence","volume":"56 9","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148261712","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}