Open set domain adaptation (OSDA) copes with the distribution and label shifts between the source and target domains simultaneously, performing accurate classification for known classes while identifying unknown class samples in the target domain. Most existing OSDA approaches, depending on the final image feature space of deep models, require manually-tuned thresholds, and may easily misclassify unknown samples as known classes. Mixture-of-experts (MoE) could be a remedy. Within an MoE, different experts handle distinct input features, producing unique expert routing patterns for various classes in a routing feature space. As a result, unknown class samples may display different expert routing patterns to known classes. This article proposes dual-space detection, which exploits the inconsistencies between the image feature space and the routing feature space to detect unknown class samples without any threshold. A graph router is further introduced to better make use of the spatial information among the image patches. Experiments on three datasets validated the effectiveness and superiority of our approach.
{"title":"Mixture-of-Experts for Open Set Domain Adaptation: A Dual-Space Detection Approach","authors":"Zhenbang Du;Jiayu An;Yunlu Tu;Jiahao Hong;Dongrui Wu","doi":"10.1109/TAI.2025.3560590","DOIUrl":"https://doi.org/10.1109/TAI.2025.3560590","url":null,"abstract":"Open set domain adaptation (OSDA) copes with the distribution and label shifts between the source and target domains simultaneously, performing accurate classification for known classes while identifying unknown class samples in the target domain. Most existing OSDA approaches, depending on the final image feature space of deep models, require manually-tuned thresholds, and may easily misclassify unknown samples as known classes. Mixture-of-experts (MoE) could be a remedy. Within an MoE, different experts handle distinct input features, producing unique expert routing patterns for various classes in a routing feature space. As a result, unknown class samples may display different expert routing patterns to known classes. This article proposes dual-space detection, which exploits the inconsistencies between the image feature space and the routing feature space to detect unknown class samples without any threshold. A graph router is further introduced to better make use of the spatial information among the image patches. Experiments on three datasets validated the effectiveness and superiority of our approach.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 12","pages":"3207-3216"},"PeriodicalIF":0.0,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145612207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-14DOI: 10.1109/TAI.2025.3560592
Guojie Li;Zhiwen Yu;Kaixiang Yang;Ziwei Fan;C. L. Philip Chen
Broad learning system (BLS) has been widely researched and applied in the field of semisupervised learning. However, current semisupervised BLS methods rely on predefined graph structures. High-dimensional small-sample data, characterized by abundant redundant and noisy features with complex distribution patterns, often leads to the construction of poor-quality predefined graphs, thereby constraining the model’s performance. Additionally, the random generation of feature and enhancement nodes in BLS, combined with limited data labels, results in suboptimal model performance. To address these issues, this article first proposes a broad learning system with adaptive locality preservation (BLS-ALP). This method employs adaptive locality preservation constraints in the output space to ensure that similar samples share the same label, iteratively updating the graph structure. To further enhance the performance of BLS-ALP, an incremental ensemble framework (IBLS-ALP) is proposed. This framework effectively mitigates the impact of redundant and noisy features by using multiple random subspaces instead of the original high-dimensional space. Additionally, IBLS-ALP enhances the utilization of a small number of labels by incorporating residual labels, thereby significantly improving the model’s overall performance. Extensive experiments conducted on various high-dimensional small-sample datasets demonstrate that IBLS-ALP exhibits superior performance.
{"title":"Incremental Semisupervised Learning With Adaptive Locality Preservation for High-Dimensional Data","authors":"Guojie Li;Zhiwen Yu;Kaixiang Yang;Ziwei Fan;C. L. Philip Chen","doi":"10.1109/TAI.2025.3560592","DOIUrl":"https://doi.org/10.1109/TAI.2025.3560592","url":null,"abstract":"Broad learning system (BLS) has been widely researched and applied in the field of semisupervised learning. However, current semisupervised BLS methods rely on predefined graph structures. High-dimensional small-sample data, characterized by abundant redundant and noisy features with complex distribution patterns, often leads to the construction of poor-quality predefined graphs, thereby constraining the model’s performance. Additionally, the random generation of feature and enhancement nodes in BLS, combined with limited data labels, results in suboptimal model performance. To address these issues, this article first proposes a broad learning system with adaptive locality preservation (BLS-ALP). This method employs adaptive locality preservation constraints in the output space to ensure that similar samples share the same label, iteratively updating the graph structure. To further enhance the performance of BLS-ALP, an incremental ensemble framework (IBLS-ALP) is proposed. This framework effectively mitigates the impact of redundant and noisy features by using multiple random subspaces instead of the original high-dimensional space. Additionally, IBLS-ALP enhances the utilization of a small number of labels by incorporating residual labels, thereby significantly improving the model’s overall performance. Extensive experiments conducted on various high-dimensional small-sample datasets demonstrate that IBLS-ALP exhibits superior performance.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 11","pages":"2990-3004"},"PeriodicalIF":0.0,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
One of the main challenges in electroencephalography (EEG) emotion recognition is the lack of understanding of the biological properties of the brain and how they relate to emotions. To address this issue, this article proposes an implicit emotion regulatory mechanism inspired contrastive learning framework (CLIER) for EEG emotion recognition. The framework simulates the complex relationship between emotions and the underlying neurobiological processes; to achieve this, the mechanism is mainly simulated through three parts. First, to leverage the interindividual variability of emotional expression, the emotion features of the individual are captured by a dynamic connection graph in the subject-dependent setting. Subsequently, reverse regulation is simulated by contrast learning based on label information and data augmentation to capture more biologically specific emotional features. Finally, caused by the asymmetry between the left and right hemispheres of the human brain in response to emotions, brain lateralization mutual learning facilitates the fusion of the hemispheres in determining emotions. Experiments on SEED, SEED-IV, SEED-V, and EREMUS datasets show impressive results: 93.4% accuracy on SEED, 90.2% on SEED-IV, 82.46% on SEED-V, and 41.63% on EREMUS. Employing an identical experimental protocol, our model demonstrated superior performance relative to the majority of existing methods, thus showcasing its effectiveness in the realm of EEG emotion recognition.
{"title":"EEG Emotion Recognition Based on an Implicit Emotion Regulatory Mechanism","authors":"Dongdong Li;Zhishuo Jin;Yujun Shen;Zhe Wang;Suo Jiang","doi":"10.1109/TAI.2025.3560593","DOIUrl":"https://doi.org/10.1109/TAI.2025.3560593","url":null,"abstract":"One of the main challenges in electroencephalography (EEG) emotion recognition is the lack of understanding of the biological properties of the brain and how they relate to emotions. To address this issue, this article proposes an implicit emotion regulatory mechanism inspired contrastive learning framework (CLIER) for EEG emotion recognition. The framework simulates the complex relationship between emotions and the underlying neurobiological processes; to achieve this, the mechanism is mainly simulated through three parts. First, to leverage the interindividual variability of emotional expression, the emotion features of the individual are captured by a dynamic connection graph in the subject-dependent setting. Subsequently, reverse regulation is simulated by contrast learning based on label information and data augmentation to capture more biologically specific emotional features. Finally, caused by the asymmetry between the left and right hemispheres of the human brain in response to emotions, brain lateralization mutual learning facilitates the fusion of the hemispheres in determining emotions. Experiments on SEED, SEED-IV, SEED-V, and EREMUS datasets show impressive results: 93.4% accuracy on SEED, 90.2% on SEED-IV, 82.46% on SEED-V, and 41.63% on EREMUS. Employing an identical experimental protocol, our model demonstrated superior performance relative to the majority of existing methods, thus showcasing its effectiveness in the realm of EEG emotion recognition.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 11","pages":"3005-3017"},"PeriodicalIF":0.0,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-11DOI: 10.1109/TAI.2025.3560248
Zhiqiang Ge;Duxin Chen;Wenwu Yu
Recently, probabilistic latent variable models have played an important role in data analytics in various industrial application scenarios, such as process monitoring, fault diagnosis, and soft sensing. Inspired by the idea of lightweight deep learning, this article proposes a new deep residual learning method for the probabilistic’ partial least squares (PLSs) model. First, layerwise probabilistic modeling is carried out to extract supervised latent variables in different hidden layers of the deep model using a well-designed expectation-maximization algorithm for parameter optimization. Through this layerwise residual learning process, more target-related latent variables can be extracted, which are supervised by the outputs of the predictive model. Next, an additional probabilistic model is constructed for information fusion and further extraction of supervised latent variables which are highly related to the modeling target. In fact, this step can be considered as an ensemble learning strategy, which has great potentials in decreasing modeling error and reducing prediction uncertainty. A soft-sensing strategy is then developed for online prediction of key variables. The performance is evaluated using two industrial examples. Compared to the shallow probabilistic model, the performance of the deep model has been improved by 10%–20%.
{"title":"Deep Residual Learning of a Probabilistic’ Partial Least Squares Model for Predictive Data Analytics","authors":"Zhiqiang Ge;Duxin Chen;Wenwu Yu","doi":"10.1109/TAI.2025.3560248","DOIUrl":"https://doi.org/10.1109/TAI.2025.3560248","url":null,"abstract":"Recently, probabilistic latent variable models have played an important role in data analytics in various industrial application scenarios, such as process monitoring, fault diagnosis, and soft sensing. Inspired by the idea of lightweight deep learning, this article proposes a new deep residual learning method for the probabilistic’ partial least squares (PLSs) model. First, layerwise probabilistic modeling is carried out to extract supervised latent variables in different hidden layers of the deep model using a well-designed expectation-maximization algorithm for parameter optimization. Through this layerwise residual learning process, more target-related latent variables can be extracted, which are supervised by the outputs of the predictive model. Next, an additional probabilistic model is constructed for information fusion and further extraction of supervised latent variables which are highly related to the modeling target. In fact, this step can be considered as an ensemble learning strategy, which has great potentials in decreasing modeling error and reducing prediction uncertainty. A soft-sensing strategy is then developed for online prediction of key variables. The performance is evaluated using two industrial examples. Compared to the shallow probabilistic model, the performance of the deep model has been improved by 10%–20%.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 11","pages":"2977-2989"},"PeriodicalIF":0.0,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145455952","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-09DOI: 10.1109/TAI.2025.3558183
Nilufar Zaman;Angshuman Jana
In today’s world, online services have revolutionized human activities and thus the consumers expect their service providers to make their online experiences more fruitful by recommending the relevant services to them. In this case, it becomes really challenging for the service providers to provide recommendation to a user whose information’s and preferences are unavailable. This issue is handled by cross-domain approach, which explores similar users across various domains in the same platform. However, the main concern with this cross-domain approach is that the information needs to be available in any domain of one platform. Thus, a multidomain recommendation is designed to optimize the recommendation system performance by analyzing the information obtained from multiple platforms. However, existing multidomain recommendation model has mainly two challenges. First, there are no overlapping users to understand the similarities between them. Second, the transfer learning approach in multidomain allows the transfer of information from only the source to the target domain. Therefore, our proposed approach consider the parallel inductive shift learning (PISL) model to address these two above-mentioned challenges. For the first challenge, we have focused to identify the similarities between user–user and user–item by considering various features of user and item. For the next challenge, our proposed model analyzes the source and the target domain simultaneously and thus does a parallel transfer of information from the source to the target domain and vice versa. We have tested our model for three real-life movie and book datasets i.e. for the movie dataset we have used Movielens, Amazon, and Netflix datasets. In contrast, for the book dataset, we have used the Amazon, Good Reads, and Book Crossing dataset, which proves to outperform the other state-of-the-art approaches.
{"title":"Parallel Inductive Shift Learning Based Recommendation System","authors":"Nilufar Zaman;Angshuman Jana","doi":"10.1109/TAI.2025.3558183","DOIUrl":"https://doi.org/10.1109/TAI.2025.3558183","url":null,"abstract":"In today’s world, online services have revolutionized human activities and thus the consumers expect their service providers to make their online experiences more fruitful by recommending the relevant services to them. In this case, it becomes really challenging for the service providers to provide recommendation to a user whose information’s and preferences are unavailable. This issue is handled by cross-domain approach, which explores similar users across various domains in the same platform. However, the main concern with this cross-domain approach is that the information needs to be available in any domain of one platform. Thus, a multidomain recommendation is designed to optimize the recommendation system performance by analyzing the information obtained from multiple platforms. However, existing multidomain recommendation model has mainly two challenges. First, there are no overlapping users to understand the similarities between them. Second, the transfer learning approach in multidomain allows the transfer of information from only the source to the target domain. Therefore, our proposed approach consider the parallel inductive shift learning (PISL) model to address these two above-mentioned challenges. For the first challenge, we have focused to identify the similarities between user–user and user–item by considering various features of user and item. For the next challenge, our proposed model analyzes the source and the target domain simultaneously and thus does a parallel transfer of information from the source to the target domain and vice versa. We have tested our model for three real-life movie and book datasets i.e. for the movie dataset we have used Movielens, Amazon, and Netflix datasets. In contrast, for the book dataset, we have used the Amazon, Good Reads, and Book Crossing dataset, which proves to outperform the other state-of-the-art approaches.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 11","pages":"2953-2965"},"PeriodicalIF":0.0,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145428951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-08DOI: 10.1109/TAI.2025.3558718
Soumyadipta Banerjee;Jiaul H. Paik
Modern deep networks are highly over-parameterized. Thus, training and testing such models in various applications are computationally intensive with excessive memory and energy requirements. Network pruning aims to find smaller subnetworks from within these dense networks that do not compromise on the test accuracy. In this article, we present a probabilistic and deterministic pruning methodology which determines the likelihood of retention of the weight parameters by modeling the layer-specific distribution of extreme values of the weights. Our method automatically finds the sparsity in each layer, unlike existing pruning techniques which require an explicit input of the sparsity information. Experiments in the present work show that deterministic–probabilistic pruning consistently achieves high sparsity levels, ranging from 65 to 95%, while maintaining comparable or improved testing accuracy across multiple datasets such as MNIST, CIFAR-10, and Tiny ImageNet, on architectures including VGG-16, ResNet-18, and ResNet-50.
{"title":"A Deterministic–Probabilistic Approach to Neural Network Pruning","authors":"Soumyadipta Banerjee;Jiaul H. Paik","doi":"10.1109/TAI.2025.3558718","DOIUrl":"https://doi.org/10.1109/TAI.2025.3558718","url":null,"abstract":"Modern deep networks are highly over-parameterized. Thus, training and testing such models in various applications are computationally intensive with excessive memory and energy requirements. Network pruning aims to find smaller subnetworks from within these dense networks that do not compromise on the test accuracy. In this article, we present a probabilistic and deterministic pruning methodology which determines the likelihood of retention of the weight parameters by modeling the layer-specific distribution of extreme values of the weights. Our method automatically finds the sparsity in each layer, unlike existing pruning techniques which require an explicit input of the sparsity information. Experiments in the present work show that deterministic–probabilistic pruning consistently achieves high sparsity levels, ranging from 65 to 95%, while maintaining comparable or improved testing accuracy across multiple datasets such as MNIST, CIFAR-10, and Tiny ImageNet, on architectures including VGG-16, ResNet-18, and ResNet-50.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"6 10","pages":"2830-2839"},"PeriodicalIF":0.0,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145196043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-30DOI: 10.1109/TAI.2025.3575036
Mingzhi Yuan;Ao Shen;Yingfan Ma;Jie Du;Qiao Huang;Manning Wang
Deep learning has significantly advanced the development of point cloud registration. However, in recent years, some methods have relied on additional sensor information or complex network designs to improve registration performance, which incurs considerable computational overhead. These methods often struggle to strike a reasonable balance between computational cost and performance gains. To address this, we propose a plug-and-play orthogonal self-ensemble module designed to enhance registration performance with minimal additional overhead. Specifically, we design a novel ensemble learning strategy to mine the complementary information within the extracted features of previous methods. Unlike most ensemble learning methods, our method does not set multiple complex models for performance enhancement. Instead, it only cascades a lightweight dual-branch network after the features extracted by the original model to obtain two sets of features with more diversity. To further reduce redundancy between features and prevent the degradation of the dual-branch network, we introduce an orthogonal constraint that ensures the features output by the two branches are more complementary. Finally, by concatenating the two sets of complementary features, the final enhanced features are obtained. Compared to the original features, these enhanced features thoroughly exploit the internal information and exhibit greater distinctiveness, leading to improved registration performance. To validate the effectiveness of our method, we plug it into GeoTransformer, resulting in consistent performance improvements across 3DMatch, KITTI, and ModelNet40 datasets. Moreover, our method is compatible with other performance-enhancing methods. In conjunction with the overlap prior in PEAL, GeoTransformer achieves a new state-of-the-art performance.
{"title":"Boosting 3-D Point Cloud Registration by Orthogonal Self-Ensemble Learning","authors":"Mingzhi Yuan;Ao Shen;Yingfan Ma;Jie Du;Qiao Huang;Manning Wang","doi":"10.1109/TAI.2025.3575036","DOIUrl":"https://doi.org/10.1109/TAI.2025.3575036","url":null,"abstract":"Deep learning has significantly advanced the development of point cloud registration. However, in recent years, some methods have relied on additional sensor information or complex network designs to improve registration performance, which incurs considerable computational overhead. These methods often struggle to strike a reasonable balance between computational cost and performance gains. To address this, we propose a plug-and-play orthogonal self-ensemble module designed to enhance registration performance with minimal additional overhead. Specifically, we design a novel ensemble learning strategy to mine the complementary information within the extracted features of previous methods. Unlike most ensemble learning methods, our method does not set multiple complex models for performance enhancement. Instead, it only cascades a lightweight dual-branch network after the features extracted by the original model to obtain two sets of features with more diversity. To further reduce redundancy between features and prevent the degradation of the dual-branch network, we introduce an orthogonal constraint that ensures the features output by the two branches are more complementary. Finally, by concatenating the two sets of complementary features, the final enhanced features are obtained. Compared to the original features, these enhanced features thoroughly exploit the internal information and exhibit greater distinctiveness, leading to improved registration performance. To validate the effectiveness of our method, we plug it into GeoTransformer, resulting in consistent performance improvements across 3DMatch, KITTI, and ModelNet40 datasets. Moreover, our method is compatible with other performance-enhancing methods. In conjunction with the overlap prior in PEAL, GeoTransformer achieves a new state-of-the-art performance.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 1","pages":"375-384"},"PeriodicalIF":0.0,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-30DOI: 10.1109/TAI.2025.3575038
Gokul Bhusal;Kevin Miller;Ekaterina Merkurjev
Active learning (AL) enhances the performance of machine learning (ML) methods, particularly in low-label rate scenarios, by judiciously selecting a limited number of unlabeled data points for labeling, with the goal of improving the performance of an underlying classifier. In this work, we introduce the multiclass AL with auction dynamics on graphs (MALADY) algorithm, which leverages an auction dynamics technique on similarity graphs for efficient AL. In particular, the proposed algorithm incorporates an AL loop using as its underlying semisupervised procedure an efficient and effective similarity graph-based auction method consisting of upper and lower bound auctions that integrate class size constraints. In addition, we introduce a novel AL acquisition function that incorporates the dual variable of the auction algorithm to measure the uncertainty in the classifier to prioritize queries near the decision boundaries between different classes. Overall, the proposed method can efficiently obtain accurate results using extremely small labeled sets containing just a few elements per class; this is crucial since labeled data are scarce for many applications. Moreover, the proposed technique can incorporate class size information, which improves accuracy even further. Last, using experiments on classification tasks and various datasets, we evaluate the performance of our proposed method and show that it exceeds that of comparison algorithms.
{"title":"MALADY: Multiclass Active Learning With Auction Dynamics on Graphs","authors":"Gokul Bhusal;Kevin Miller;Ekaterina Merkurjev","doi":"10.1109/TAI.2025.3575038","DOIUrl":"https://doi.org/10.1109/TAI.2025.3575038","url":null,"abstract":"Active learning (AL) enhances the performance of machine learning (ML) methods, particularly in low-label rate scenarios, by judiciously selecting a limited number of unlabeled data points for labeling, with the goal of improving the performance of an underlying classifier. In this work, we introduce the multiclass AL with auction dynamics on graphs (MALADY) algorithm, which leverages an auction dynamics technique on similarity graphs for efficient AL. In particular, the proposed algorithm incorporates an AL loop using as its underlying semisupervised procedure an efficient and effective similarity graph-based auction method consisting of upper and lower bound auctions that integrate class size constraints. In addition, we introduce a novel AL acquisition function that incorporates the dual variable of the auction algorithm to measure the uncertainty in the classifier to prioritize queries near the decision boundaries between different classes. Overall, the proposed method can efficiently obtain accurate results using extremely small labeled sets containing just a few elements per class; this is crucial since labeled data are scarce for many applications. Moreover, the proposed technique can incorporate class size information, which improves accuracy even further. Last, using experiments on classification tasks and various datasets, we evaluate the performance of our proposed method and show that it exceeds that of comparison algorithms.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 1","pages":"385-398"},"PeriodicalIF":0.0,"publicationDate":"2025-03-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898218","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article investigates traffic data cognitive modelling problem in real traffic scene by fully utilizing multiscale spatio-temporal dependence between multiple traffic nodes, along with a novel dynamic graph convolutional network (GCN). Most recently, the deep learning network model is weighed down by some practical problems focused on as follows: 1) The existing graph convolution operations typically aggregate information from the given k-hop neighbors; and 2) How to model the similarity of traffic data patterns among these nodes given the spatio-temporal heterogeneity of traffic data. In this article, we propose a novel hierarchical traffic data cognitive modelling framework called multiscale spatio-temporal dynamic graph convolutional network architecture (MSST-DGCN). And, a multiscale graph convolution module is first constructed to expand the receptive field of convolutional operations, by developing a novel sub-GCNs cumulative concatenation mechanism. Meanwhile, two specified dynamic graphs are designed to model the spatio-temporal correlation among these nodes from both a proximity and long-term perspective through a novel Gaussian calculation strategy, which are efficiently able to represent/cognize the dynamic similarity of traffic data patterns. Through a series of qualitative evaluations, the present model has the ability to perceive the traffic data pattern states of nodes. At last, two real world traffic datasets experiments are developed to show that the proposed approach achieves state-of-the-art traffic data cognitive performance.
{"title":"A Novel Multiscale Dynamic Graph Convolutional Network for Traffic Data Cognition","authors":"Jiyao An;Zhaohui Pu;Qingqin Liu;Lei Zhang;Md Sohel Rana","doi":"10.1109/TAI.2025.3574655","DOIUrl":"https://doi.org/10.1109/TAI.2025.3574655","url":null,"abstract":"This article investigates traffic data cognitive modelling problem in real traffic scene by fully utilizing multiscale spatio-temporal dependence between multiple traffic nodes, along with a novel dynamic graph convolutional network (GCN). Most recently, the deep learning network model is weighed down by some practical problems focused on as follows: 1) The existing graph convolution operations typically aggregate information from the given k-hop neighbors; and 2) How to model the similarity of traffic data patterns among these nodes given the spatio-temporal heterogeneity of traffic data. In this article, we propose a novel hierarchical traffic data cognitive modelling framework called multiscale spatio-temporal dynamic graph convolutional network architecture (MSST-DGCN). And, a multiscale graph convolution module is first constructed to expand the receptive field of convolutional operations, by developing a novel sub-GCNs cumulative concatenation mechanism. Meanwhile, two specified dynamic graphs are designed to model the spatio-temporal correlation among these nodes from both a proximity and long-term perspective through a novel Gaussian calculation strategy, which are efficiently able to represent/cognize the dynamic similarity of traffic data patterns. Through a series of qualitative evaluations, the present model has the ability to perceive the traffic data pattern states of nodes. At last, two real world traffic datasets experiments are developed to show that the proposed approach achieves state-of-the-art traffic data cognitive performance.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":"7 1","pages":"362-374"},"PeriodicalIF":0.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145898231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：