检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年第1期6卷 221-233页

作者： Yang, Yang Wang, Chao Gong, Lei Wu, Min Chen, Zhenghua Li, Xiang Chen, Xianglan Zhou, Xuehai University of Science and Technology of China School of Computer Science and Technology Hefei230026 China 138632 Singapore East China Normal University School of Data Science and Engineering Shanghai200062 China

Ensemble object detectors have demonstrated remarkable effectiveness in enhancing prediction accuracy and uncertainty quantification. However, their widespread adoption is hindered by significant computational and storage demands, limiting their feasibility in resource-constrained settings. To overcome this, researchers have focused on distilling the knowledge from ensemble object detectors into a single model. In this article, we introduce probabilization based ensemble distillation (ProbED), an innovative ensemble distillation framework that consolidates knowledge from multiple object detectors into a single, resource-efficient model. Unlike traditional ensemble distillation methods that average the outputs of subteachers, ProbED captures comprehensive outcome distributions from all subteachers, providing a more nuanced approach to knowledge transfer. ProbED employs knowledge probabilization to achieve a sophisticated and refined aggregation of teacher knowledge, including feature knowledge, semantic knowledge, and localization knowledge, resulting in dual improvements in prediction accuracy and uncertainty quantification for the student model. In particular, ProED's novel knowledge probabilization-based approach to aggregating teacher knowledge is inspired by our empirical observations, which demonstrate that knowledge probabilization excels in effectively representing uncertainty, improving prediction, and facilitating robust knowledge transfer. Furthermore, we introduce a random smoothing perturbation technique to modify inputs within ProbED, further enhancing the distillation process. Extensive experiments highlight ProbED's ability to significantly enhance the prediction accuracy and uncertainty quantification of various object detectors, demonstrating its superior performance compared to other state-of-the-art techniques. © 2024 IEEE.

关键词： Teaching

来源：评论

学校读者我要写书评

暂无评论

HGTFM: Hierarchical Gating-Driven Transformer Fusion Model for Robust Multimodal Sentiment Analysis

引用

IEEE Access 2025年 13卷 74430-74445页

作者： Yang, Chengcheng Liang, Zhiyao Yan, Dashun Hu, Zeng Wu, Ting Macau University of Science and Technology School of Computer Science and Engineering Avenida Wai Long Taipa China Zhongkai University of Agriculture and Engineering School of Information Science and Technology Guangzhou510550 China

In multimodal sentiment analysis, a significant challenge lies in quantifying the contribution of each modality and achieving effective modality fusion. This paper presents a Hierarchical Gating-Driven Transformer Fusion Model (HGTFM), which effectively achieves multimodal data fusion through an advanced transformer architecture and a Hierarchical Gated Fusion Module (HGFM). The model first encodes audio and video features using LSTM and CNN, respectively, while employing BERT to encode text features. This approach effectively enhances the model ability to manage long-range temporal dependencies. Furthermore, the model introduces HGFM, which dynamically integrates multimodal features across different levels after processing by the modality interaction network (MIN), thus intelligently adjusting the contribution weights of each modality. In particular, the model incorporates a Unimodal Label Generation Module (ULGM), which considers the differences between the modalities. This enables HGFM to more accurately identify the contributions of samples with significant modal variations, thus optimizing the module potential. The experimental results on several established benchmark datasets demonstrate that HGTFM achieves comparable performance improvements in multimodal sentiment analysis tasks. © 2013 IEEE.

关键词： Sentiment analysis Transformers Analytical models Data models Adaptation models Tensors Accuracy Noise measurement Training Robustness

来源：评论

学校读者我要写书评

暂无评论

A two-stage method with twin autoencoders for the degradation trajectories prediction of lithium-ion batteries

引用

Journal of Energy Chemistry 2025年第4期103卷 759-772页

作者： Lei Cai Jing Yan Haiyan Jin Jinhao Meng Jichang Peng Bin Wang Wei Liang Remus Teodorescu School of Computer Science and Engineering Xi'an University of Technology School of Electrical Engineering Xi'an Jiaotong University Smart Grid Research Institute Nanjing Institute of Technology Department of Energy Aalborg University

To predict the lithium-ion(Li-ion) battery degradation trajectory in the early phase,arranging the maintenance of battery energy storage systems is of great ***,under different operation conditions,Li-ion batteries present distinct degradation patterns,and it is challenging to capture negligible capacity fade in early *** the data-driven method showing promising performance,insufficient data is still a big issue since the ageing experiments on the batteries are too slow and *** this study,we proposed twin autoencoders integrated into a two-stage method to predict the early cycles' degradation *** two-stage method can properly predict the degradation from course to *** twin autoencoders serve as a feature extractor and a synthetic data generator,***,a learning procedure based on the long-short term memory(LSTM) network is designed to hybridize the learning process between the real and synthetic *** performance of the proposed method is verified on three datasets,and the experimental results show that the proposed method can achieve accurate predictions compared to its competitors.

关键词： Autoencoder Battery degradation trajectory Early prediction Synthetic data generation

来源：评论

学校读者我要写书评

暂无评论

Robust Learning under Hybrid Noise

引用

ACM Transactions on Intelligent Systems and technology 2025年第2期16卷 1-27页

作者： Wei, Yang Chen, Shuo Ye, Shanshan Han, Bo Gong, Chen School of Computer Science Nanjing University of Science and Technology Nanjing China Riken Aip Chuo-ku Japan University of Technology Sydney Broadway Australia Hong Kong Baptist University Hong Kong Hong Kong School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

Feature noise and label noise are ubiquitous in practical scenarios, which pose great challenges for training a robust machine learning model. Most previous approaches usually deal with only a single problem of either feature noise or label noise. However, in real-world applications, hybrid noise, which contains both feature noise and label noise, is very common due to the unreliable data collection and annotation processes. Although some results have been achieved by a few representation learning based attempts, this issue is still far from being addressed with promising performance and guaranteed theoretical analyses. To address the challenge, we propose a novel unified learning framework called Feature and Label Recovery (FLR) to combat the hybrid noise from the perspective of data recovery, where we concurrently reconstruct both the feature matrix and the label matrix of input data. Specifically, the clean feature matrix is discovered by the low-rank approximation, and the ground-truth label matrix is embedded based on the recovered features with a nuclear norm regularization. Meanwhile, the feature noise and label noise are characterized by their respective adaptive matrix norms to satisfy the corresponding maximum likelihood. As this framework leads to a non-convex optimization problem, we develop the non-convex Alternating Direction Method of Multipliers (ADMM) with the convergence guarantee to solve our learning objective. We also provide the theoretical analysis to show that the generalization error of FLR can be upper-bounded in the presence of hybrid noise. Experimental results on several typical benchmark datasets clearly demonstrate the superiority of our proposed method over the state-of-the-art robust learning approaches for various noises. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Entity and Relationship Extraction with High-quality Spans and Enhanced Marker Strategies

引用

Machine Intelligence Research 2025年第2期22卷 289-303页

作者： Qibin Li Nianmin Yao Nai Zhou School of Computer Science Dalian University of TechnologyDalian 116024China Ningbo Institute of Dalian University of Technology Ningbo 116026China

Entity and relation extraction is a critical task in information *** approaches have emphasized obtaining improved span ***,existing work suffers from two major ***,there is an overabundance of low-quality candidate spans,which hinders the effective extraction of information from high-quality candidate ***,the information encoded by existing marker strategies is often too simple to fully capture the nuances of the span,resulting in the loss of potentially valuable *** address these issues,we propose an enhancing entity and relation extraction with high-quality spans and enhanced marker(HSEM)strategies,it assigns adaptive weights to different spans in order to make the model more focused on high quality ***,the HSEM model enriches marker representation to incorporate more span information and enhance entity ***,we design a span scoring framework that assesses span quality based on the fusion of internal information and focuses the model on training high-quality samples to improve *** results on six benchmark datasets demonstrate that our model achieves state-of-the-art results after discriminating span quality.

关键词： High-quality spans enhanced marker span scoring framework span representations entity and relation extraction

来源：评论

学校读者我要写书评

暂无评论

DualRec: A Collaborative Training Framework for Device and Cloud Recommendation Models

引用

IEEE Transactions on Mobile Computing 2025年第6期24卷 5202-5213页

作者： Zhang, Ye Deng, Yongheng Yue, Sheng Li, Qiushi Ren, Ju Beijing Information Science and Technology University School of Computer Science Beijing100101 China Tsinghua University Department of Computer Science and Technology BNRist Beijing100084 China Zhongguancun Laboratory Beijing100084 China

Recommendation systems (RS) play a vital role in various domains. However, under recent data regulations like General Data Protection Regulation (GDPR), traditional RS that rely on collecting user's interaction data centrally face significant challenges. Federated learning (FL) enables collaborative model training among users while keeping their private data locally. Yet, the constrained resources of devices often limit the size of the learned model, resulting in suboptimal recommendation performance. To overcome the dilemma of data accessibility and model size, we propose DualRec, a novel collaborative training framework for device and cloud recommendation models. In DualRec, users train lightweight models on devices to harness their local private data, while a larger model is simultaneously trained on the cloud server to exploit its substantial resources. Devices and the cloud server collaboratively train their models, compensating for individual limitations of model size and data availability, enabling mutual empowerment and benefits. Specifically, we introduce an efficient aggregation mechanism for recommendation models to boost the collaborative training performance of device models. With the learned device models, we propose to generate pseudo user interaction data to train the server model. To enhance the training performance of the server model, we design an automated denoising mechanism to mitigate the negative impact of noisy samples in the generated pseudo dataset. Finally, the learned knowledge of the server model is distilled to device models for enhanced on-device recommendation performance. Extensive experiments demonstrate the superior performance of DualRec compared to state-of-the-art baselines. © 2002-2012 IEEE.

关键词： Collaborative learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing aviation safety: Machine learning for real-time ADS-B injection detection through advanced data analysis

引用

Alexandria Engineering Journal 2025年 126卷 262-276页

作者： Rahman, Md. Atiqur Bhuiyan, Touhid Ali, M. Ameer Department of Computer Science and Engineering East West university Dhaka Bangladesh School of Information Technology Washington University of Science and Technology Alexandria United States Department of Computer Science and Engineering Bangladesh University of Business and Technology Dhaka Bangladesh

Airplanes play a critical role in global transportation, ensuring the efficient movement of people and goods. Although generally safe, aviation systems occasionally encounter incidents and accidents that underscore the need for proactive risk management. This study employs machine learning to detect abnormalities in commercial aircraft operations using Automatic Dependent Surveillance–Broadcast (ADS-B) data. Given the growing reliance on ADS-B technology, concerns regarding its susceptibility to security breaches, such as injection attacks, have intensified. To address these vulnerabilities, we propose a robust ADS-B injection detection system. Employing GridSearchCV for model optimization, it effectively identifies and categorizes injection risks. The system's performance, evaluated using the ADS-B Message Injection Attacks Dataset, achieves outstanding results, including a value of 0.9970 for the accuracy, precision, recall, and F1 score. The proposed classifier also demonstrates a higher area under the curve (0.9999), specificity (0.9956), and Cohen's kappa (0.9954) than existing approaches, while achieving a lower log loss (0.0107). This research significantly enhances aviation security by introducing a highly accurate, computationally efficient, and reliable real-time detection model for ADS-B injection attacks, ensuring the integrity and resilience of modern flight control systems. © 2025 The Authors

关键词： Risk management

来源：评论

学校读者我要写书评

暂无评论

Robust Detection of Malicious Encrypted Traffic via Contrastive Learning

引用

IEEE Transactions on Information Forensics and Security 2025年 20卷 4228-4242页

作者： Shen, Meng Wu, Jinhe Ye, Ke Xu, Ke Xiong, Gang Zhu, Liehuang Beijing Institute of Technology School of Cyberspace Science and Technology Beijing100081 China Beijing Institute of Technology School of Computer Science Beijing100081 China Tsinghua University Department of Computer Science and Technology Beijing100084 China Institute of Information Engineering Chinese Academy of Sciences Beijing100190 China

Traffic encryption is widely used to protect communication privacy but is increasingly exploited by attackers to conceal malicious activities. Existing malicious encrypted traffic detection methods rely on large amounts of labeled samples for training, limiting their ability to quickly respond to new attacks. These methods also are vulnerable to traffic obfuscation strategies, such as injecting dummy packets. In this paper, we propose SmartDetector, a robust malicious encrypted traffic detection method via contrastive learning. We first propose a novel traffic representation named Semantic Attribute Matrix (SAM), which can effectively distinguish between malicious and benign traffic. We also design a data augmentation method to generate diverse traffic samples, which makes the detection model more robust against different traffic obfuscation strategies. We propose a malicious encrypted traffic classifier that first pre-trains a model via contrastive learning to learn deep representations from unlabeled data, then fine-tunes the model with a supervised classifier to achieve accurate detection even with only a few labeled samples. We conduct extensive experiments with five public datasets to evaluate the performance of SmartDetector. The results demonstrate that it outperforms the state-of-the-art (SOTA) methods in three typical scenarios. Specifically, in the evasion attack detection scenario, SmartDetector achieves an F1 score and AUC above 93%, with average improvements of 19.84% and 18.17% over the SOTA method, respectively. © 2005-2012 IEEE.

关键词： Data obfuscation

来源：评论

学校读者我要写书评

暂无评论

Characterizing the app recommendation relationships in the iOS app store: a complex network's perspective

引用

science China(Information sciences) 2025年第4期68卷 284-299页

作者： Gang HUANG Fuqi LIN Yun MA Haoyu WANG Qingxiang WANG Gareth TYSON Xuanzhe LIU Key Lab of High Confidence Software Technologies (Peking University) Ministry of Education School of Computer Science Peking University Institute for Artificial Intelligence Peking University School of Cyber Science and Engineering Huazhong University of Science and Technology Alibaba Group Hong Kong University of Science and Technology (Guangzhou)

Mobile apps have become widely adopted in our daily lives. To facilitate app discovery, most app markets provide recommendations for users, which may significantly impact how apps are accessed. However, little has been known about the underlying relationships and how they reflect(or affect) user behaviors. To fill this gap, we characterize the app recommendation relationships in the i OS app store from the perspective of the complex network. We collect a dataset containing over 1.3 million apps and 50 million app recommendations. This dataset enables us to construct a complex network that captures app recommendation relationships. Through this, we explore the recommendation relationships between mobile apps and how these relationships reflect or affect user behavior patterns. The insights gained from our research can be valuable for understanding typical user behaviors and identifying potential policy-violating apps.

关键词： mobile app recommendation complex network user behavior policy-violating app

来源：评论

学校读者我要写书评

暂无评论

Gender opposition recognition method fusing emojis and multi-features in Chinese speech

引用

Soft Computing 2025年第4期29卷 2379-2390页

作者： Zhang, Shunxiang Ma, Zichen Li, Hanchen Liu, Yunduo Chen, Lei Li, Kuan-Ching School of Computer Science and Engineering Anhui University of Science and Technology Anhui Huainan China School of Computer Huainan Normal University Anhui Huainan China Providence University Taichung Taiwan

Speech with gender opposition on the internet have been causing antagonism, gamophobia, and pregnancy phobia among young groups. Recognizing gender opposition speech contributes to maintaining a healthy online environment and security in cyberspace. Traditional recognition model ignores the Chinese-owned features and emojis, which inevitably affects the recognition accuracy of gender opposition. To tackle this issue, a gender opposition recognition method fusing emojis and multi-features in Chinese speech(GOR-CS) is proposed. Firstly, the exBERT method is employed to expand the encoding of emojis into the BERT vocabulary, which can ensure BERT to extract the basis vectors containing characters and emojis information. Then, the feature vectors containing Wubi, Zhengma, and Pinyin information are extracted by Word2Vec to obtain the Chinese-owned features of gender opposition text. Further, the proposed basis vector and feature vectors are fused and then fed into the Bi-GRU network to extract deeper semantics from input sentences. Finally, to determine whether the speech are related to gender opposition, the sentiment polarities are calculated with the fully connected layer and SoftMax function. Experimental results show that the proposed method can effectively improve the accuracy of gender opposition recognition. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：