IEEE Transactions on Big Data最新文献_第3页

OpenNOD: Modeling Event-Driven Opinion Dynamics in Open Social Networks OpenNOD：在开放社会网络中建模事件驱动的意见动态

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-06-01 Epub Date: 2025-12-31 DOI: 10.1109/TBDATA.2025.3650059

Haihua Xie;Miao He;Xiaoming Zhang;Lin Liu;Philip S. Yu

{"title":"OpenNOD: Modeling Event-Driven Opinion Dynamics in Open Social Networks","authors":"Haihua Xie;Miao He;Xiaoming Zhang;Lin Liu;Philip S. Yu","doi":"10.1109/TBDATA.2025.3650059","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3650059","url":null,"abstract":"Conventional opinion dynamics models primarily focus on interpersonal interactions within closed environments, often overlooking the substantial influence of external events. However, in open social networks, public opinion is highly dynamic, frequently shifting in response to rapidly evolving and unpredictable external stimuli. To address these limitations, this paper proposes <bold>OpenNOD</b> (Open Network Opinion Dynamics), a novel framework for modeling event-driven opinion dynamics in open social environments. Grounded in real-world social media data, OpenNOD introduces a group-level opinion dynamics model that captures both intergroup influences among opinion communities and the impact of external events. To support this framework, two key technical modules are introduced: (1) As online opinions are primarily expressed through natural language, an in-context learning-based method is designed for individual opinion classification. This approach enhances opinion extraction accuracy by leveraging class-aware contrastive similarity to select optimal demonstration examples for large language models (LLMs); (2) To quantify the influence of external events over time, a temporal event influence measurement module is designed to integrate both the intensity of events and their directional impact on public sentiment. By jointly modeling internal and external factors, OpenNOD enables more realistic modeling of how real-world events trigger rapid shifts in public opinion and reshape influence structures. Extensive experiments on two real-world datasets demonstrate the framework’s adaptability across diverse topics and its effectiveness in capturing nonlinear opinion dynamics. These findings highlight OpenNOD’s potential to bridge the gap between traditional modeling and the complex behaviors observed on modern online platforms, offering a robust foundation for future studies.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"921-935"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148507798","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

Substructure-Aware Graph Representation Learning via Adaptive Node Absorbing 基于自适应节点吸收的子结构感知图表示学习

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-06-01 Epub Date: 2025-12-24 DOI: 10.1109/TBDATA.2025.3648304

Ning Liu;Songlei Jian;Dongsheng Li

{"title":"Substructure-Aware Graph Representation Learning via Adaptive Node Absorbing","authors":"Ning Liu;Songlei Jian;Dongsheng Li","doi":"10.1109/TBDATA.2025.3648304","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3648304","url":null,"abstract":"As the pivot to understand the property of the whole graph, substructure extraction plays an essential role in graph representation learning. Following a “convolution + pooling” paradigm, recent progress mainly relies on cluster pooling to achieve substructure compression. However, it suffers from high complexity caused by dense implementation and poor effectiveness resulting from rigid node assignment. Besides, the separating design of the universally adopted paradigm leads to substantial redundancy of model architecture and underutilization of convolution capability. To overcome the above dilemmas, in this paper, we propose GAS, a sparsely implemented graph representation learning model that can adaptively discover critical substructures in a flexible manner through a simplified paradigm. Specifically, GAS reformulates substructure construction from a dense node assignment problem to a sparse node absorbing problem. Besides, we propose an adaptive absorbing strategy to overcome the suboptimality caused by rigidity. It allows multi-substructure compatibility for high-influential nodes while single-substructure qualification for low-influential nodes, making a leap in the flexible division of substructures with better local clarity. In addition,we design AConv layer, which empowers convolution operation with substructure construction capability, leading to a simplified graph representation learning paradigm without extra pooling layers. Extensive experiments on graph classification and molecular property prediction tasks demonstrate a significant improvement of GAS with an average performance gain of 8.77% and 4.65% compared with SOTA competitors. Investigation of parameter size and training time quantifies the superiority of GAS with high efficiency and tractability for large-scale graphs.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"855-868"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148508821","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

MS2M: Multi-Granularity Self-Supervised Second-Order Multiple Instance Learning for Breast Cancer Pathology Image MS2M：乳腺癌病理图像的多粒度自监督二阶多实例学习

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-06-01 Epub Date: 2025-12-24 DOI: 10.1109/TBDATA.2025.3648289

Zhenwei Wang;Yiming Sun;Haitao Yao;Guangjie Han;Bingcai Chen;Pengfei Wang;Jianxin Zhang

{"title":"MS2M: Multi-Granularity Self-Supervised Second-Order Multiple Instance Learning for Breast Cancer Pathology Image","authors":"Zhenwei Wang;Yiming Sun;Haitao Yao;Guangjie Han;Bingcai Chen;Pengfei Wang;Jianxin Zhang","doi":"10.1109/TBDATA.2025.3648289","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3648289","url":null,"abstract":"Combining Big Data and deep learning can analyze large-scale breast cancer pathology images for auxiliary diagnosis. Furthermore, Whole Slide Images (WSIs) of breast cancer pathology offer detailed tissue feature information, which supports the accurate identification of malignant lesions. Current approaches combine Self-Supervised Learning (SSL) and Multiple Instance Learning (MIL) for WSI analysis, aiming to address the issues of billion-level pixels in a single WSI and the lack of precise annotations. However, pseudo-labels produced by SSL frequently lack accuracy, and MIL fails to effectively integrate global information at the WSI level, resulting in performance bottlenecks. This paper proposes the Multi-granularity Self-supervised Second-order MIL (MS2M) to tackle these issues. MS2M first achieves instance-level fine-grained feature learning through multi-granularity SSL and optimizes instance-level representations using bag-level labels within the MIL framework. Then, the transformer captures long-range dependencies between instances. When combined with second-order (covariance) pooling, it also captures high-order relational information. This process generates a robust bag-level representation. MS2M achieves accuracies of 0.9845 and 0.9719 on the CAMELYON16 and private breast cancer WSI datasets, respectively, outperforming existing methods.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"869-880"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148505879","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

rPBFT: Reliable Practical Byzantine Fault Tolerance Mechanism for Faulty Distributed Networks 故障分布式网络的可靠实用拜占庭容错机制

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-06-01 Epub Date: 2025-12-15 DOI: 10.1109/TBDATA.2025.3644587

Eunyoung Lee;Namryeong Kim;Chaerim Han;Nayeon Shin;Ilgu Lee

{"title":"rPBFT: Reliable Practical Byzantine Fault Tolerance Mechanism for Faulty Distributed Networks","authors":"Eunyoung Lee;Namryeong Kim;Chaerim Han;Nayeon Shin;Ilgu Lee","doi":"10.1109/TBDATA.2025.3644587","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3644587","url":null,"abstract":"The adoption of blockchain technology across financial, industrial, and IoT ecosystems has increased the demand for reliable and scalable consensus mechanisms. Traditional Practical Byzantine Fault Tolerance (PBFT) offers strong consistency and finality but scales poorly due to high message complexity and communication overhead. To address this, we propose Reliable Practical Byzantine Fault Tolerance (rPBFT), a consensus mechanism that improves performance and fault tolerance in large-scale distributed systems. rPBFT evaluates node reliability through message exchanges during consensus and dynamically forms a committee based on reliability scores. This reduces communication load, improves scalability, and prevents performance degradation caused by faulty nodes while maintaining decentralization without representative election. Experiments evaluated rPBFT against PBFT and HotStuff. In fault-free environments, PBFT achieved better performance due to the absence of reliability computation overhead. Under faulty conditions, rPBFT outperformed PBFT and HotStuff, achieving 30%–40% lower latency, 50%–80% higher throughput, and a 33% improvement in consensus success rate. In fault-tolerance tests, rPBFT maintained stable throughput as the Byzantine ratio and network fault ratio increased, demonstrating resilience via a reliability-based recovery model. These results confirm that rPBFT enhances scalability and robustness while preserving the core properties of Byzantine fault-tolerant consensus, making it well-suited for large-scale blockchain and IoT environments.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"814-828"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11300942","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148505811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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}

引用次数: 0

Aligning Authentic Location Shares With Mobility Information Bottleneck 基于移动信息瓶颈的真实位置共享

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-06-01 Epub Date: 2026-01-13 DOI: 10.1109/TBDATA.2026.3653638

Qiang Gao;Letian Ning;Li Huang;Chaoran Liu;Xueqin Chen;Fan Zhou

{"title":"Aligning Authentic Location Shares With Mobility Information Bottleneck","authors":"Qiang Gao;Letian Ning;Li Huang;Chaoran Liu;Xueqin Chen;Fan Zhou","doi":"10.1109/TBDATA.2026.3653638","DOIUrl":"https://doi.org/10.1109/TBDATA.2026.3653638","url":null,"abstract":"The rise of location-sharing applications (LSA) marks a significant convergence between social networks and Geographic Information Systems (GIS). With a large number of users, LSAs generate vast amounts of location-sharing data every day. Analyzing these data can enhance our understanding of human mobility behavior and drive the development of various mobility-aware downstream tasks. However, due to the unintentional acts driven by personal fuzzy check-ins and the attraction incentives intentionally offered by LSAs, location-sharing data often exhibits uncertainties, i.e., (purposely) fake shares exist in user trajectories, further affecting reliability/authenticity. Thus, inferring the real location shares in a user’s trajectory becomes an urgent task. In this study, we formalize this challenge as Authentic Location-Share Alignment (AuLA) and correspondingly propose the AuLAB solution to address it. Specifically, AuLAB is built upon a newly derived Mobility Information Bottleneck (MIB) mechanism, grounded in Information Bottleneck (IB) theory, to enhance mobility representation learning for aligning authentic shares. It also introduces context-aware attention to capture global and local dependencies among location shares, ensuring a comprehensive understanding of interactive dynamics. Experiments on four real-world datasets demonstrate the superiority of our proposed AuLAB over existing methods.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 3","pages":"977-993"},"PeriodicalIF":8.1,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148506439","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

Toward Federated Learning Against Noisy Clients via CLIP-Guided Prototypes 基于clip引导原型的针对嘈杂客户端的联合学习

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-04-01 Epub Date: 2025-11-10 DOI: 10.1109/TBDATA.2025.3630997

Zhou Tan;De Li;Jianping Cai;Yirui Huang;Chunpei Li;Ximeng Liu

{"title":"Toward Federated Learning Against Noisy Clients via CLIP-Guided Prototypes","authors":"Zhou Tan;De Li;Jianping Cai;Yirui Huang;Chunpei Li;Ximeng Liu","doi":"10.1109/TBDATA.2025.3630997","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3630997","url":null,"abstract":"Federated Noisy Labels Learning (FNLL) allows global model to be jointly trained on multiple clients with varying degrees of noisy labels while preserving privacy, and despite recent research advances, distinguishing between client clean and noisy samples is still tricky since the distribution of labels among clients is always both noisy and class-imbalanced, leading to the poor performance of existing FNLL methods. To address this problem, we propose a novel framework called FedPN, the first framework to utilize Contrastive Language-Image Pre-training (CLIP) for federated noisy labels tasks. Then, to achieve higher performance for the global model, we introduce an attention-based Prototype Adapter to identify more plausible local data for local model training, further improving training stability. We validate the effectiveness of FedPN by conducting extensive experiments on benchmark datasets under both Independently and Identically Distributed (IID) and Non-IID data partitions. The experimental results show that FedPN can effectively filter noisy samples from different clients, and compared with the state-of-the-art FNLL method, the FedPN achieves at most and at least 8.39% and 0.88% performance improvement in the case of highly heterogeneous noisy labels.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 2","pages":"509-524"},"PeriodicalIF":8.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148506321","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

Long-Tail Inductive Feature Transfer for Knowledge-Aware Multi-Behavior Recommendation 面向知识感知多行为推荐的长尾归纳特征迁移

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-04-01 Epub Date: 2025-10-14 DOI: 10.1109/TBDATA.2025.3621152

Cheng Li;Yong Xu;Suhua Tang;Weiguo Wang;Xin He;Jinde Cao

{"title":"Long-Tail Inductive Feature Transfer for Knowledge-Aware Multi-Behavior Recommendation","authors":"Cheng Li;Yong Xu;Suhua Tang;Weiguo Wang;Xin He;Jinde Cao","doi":"10.1109/TBDATA.2025.3621152","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3621152","url":null,"abstract":"Multi-Behavior recommendation captures fine- grained user intents by jointly modeling multiple interaction behaviors, significantly improving recommendation accuracy and diversity. However, existing methods often overlook complex semantic relations between items and entities. Moreover, constructing separate subgraphs for different interaction types frequently results in sparse graph structures, which exacerbates the long-tail problem. To address these challenges, we propose Long-tail Inductive Feature Transfer (LIFT), a method for knowledge-aware multi-behavior recommendation. To mitigate uneven node distribution in graph structures, we introduce a knowledge transfer-based feature reconstruction mechanism. Specifically, we first drop a portion of the neighbors of head nodes to construct proxy representations for tail nodes, training a reconstructor on the tail proxy to reconstruct the original head node features. The trained reconstructor is then used to backfill missing neighbor information for tail nodes, thereby achieving a more balanced feature distribution across nodes. Furthermore, we integrate multi-behavior and semantic contrastive learning to jointly optimize the representations. Extensive experiments on four datasets demonstrate that LIFT outperforms state-of-the-art methods, with further analysis validating its uniformity in representation learning.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 2","pages":"362-375"},"PeriodicalIF":8.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148504870","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

Incremental Granular-Ball Fuzzy Concept Learning With Information Filtering and Fusion 基于信息过滤和融合的增量颗粒球模糊概念学习

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-04-01 Epub Date: 2025-12-03 DOI: 10.1109/TBDATA.2025.3639902

Keyi Guo;Jinhai Li;Shuyin Xia

{"title":"Incremental Granular-Ball Fuzzy Concept Learning With Information Filtering and Fusion","authors":"Keyi Guo;Jinhai Li;Shuyin Xia","doi":"10.1109/TBDATA.2025.3639902","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3639902","url":null,"abstract":"Concept-cognitive learning (CCL) simulates the human brain by incrementally adjusting the concept space to adapt to the dynamic environment. Traditional CCL methods use fuzzy granular concepts to handle continuous data, but the uncertainty caused by information granularity to form fuzzy cognitive operators may lead to low cognitive accuracy. Consequently, we propose a granular-ball fuzzy concept incremental learning (GBFCIL) method to obtain the optimal initial fuzzy concept space by granular-ball computing (GBC) while utilizing purity filtering mechanism to enhance GBC’s responsiveness on dynamic data. Specifically, we introduce the definition of granular-ball fuzzy concept via the granular-ball knowledge representation method. And then, we build a static-dynamic balanced framework for constructing granular-ball fuzzy concept space: (1) an optimal initial granular-ball fuzzy concept space is established by GBC to achieve static perception of data; (2) the updating process of dynamic data involves internal generation among new samples and external fusion between the previous granular-ball fuzzy concept space and the newly learned concepts after filtering. Subsequently, a novel GBFCIL-based classification model is developed, along with the analysis of the optimal radius. Finally, experiments verify the advancement of GBFCIL in classification performance and robustness.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 2","pages":"568-583"},"PeriodicalIF":8.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148505284","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

Weakly-Supervised Camouflaged Object Detection via SAM-Guided Resolution Iteration Learning 基于sam引导的分辨率迭代学习的弱监督伪装目标检测

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-04-01 Epub Date: 2025-10-23 DOI: 10.1109/TBDATA.2025.3624975

Yanliang Ge;Yuxi Zhong;Qiao Zhang;Hongbo Bi;Tian-Zhu Xiang

{"title":"Weakly-Supervised Camouflaged Object Detection via SAM-Guided Resolution Iteration Learning","authors":"Yanliang Ge;Yuxi Zhong;Qiao Zhang;Hongbo Bi;Tian-Zhu Xiang","doi":"10.1109/TBDATA.2025.3624975","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3624975","url":null,"abstract":"Weakly supervised camouflaged object detection (WS-COD) aims to address the critical task of identifying visually assimilated objects concealed within heterogeneous backgrounds under sparse supervisory signals. However, current WS-COD frameworks suffer from compromised structural integrity, stemming from cross-hierarchical feature discrepancy and constrained cross-level information flow, which induces structural misalignment and context fragmentation in multi-granularity feature fusion. To overcome the limitation, we propose a novel SAM-guided Resolution Iteration Learning Network (SAM-RNet) that synergizes foundation model priors with multi-resolution feature refinement. Our technical contributions are threefold: (1) We utilize the Segment Anything Model (SAM) to produce high-quality masks, effectively mitigating supervision insufficiency through large-scale visual knowledge distillation. (2) We design a resolution iteration mechanism where high-resolution features progressively refine low-resolution counterparts through an Interactive Refinement Module (IRM) - a dual-branch architecture enabling hierarchical feature interaction and enhancement through branch collaboration and attention mechanism, complemented by an iterative feedback loss to enforce multi-scale feature learning. (3) We develop a Decoder with cross-layer fusion operations, enabling the aggregation of features from object and background contexts for precise object segmentation. Finally, extensive experiments demonstrate that SAM-RNet is superior to existing WS-COD methods across three COD datasets, achieving average improvements of 4.37%, 4.60%, 7.00%, and 24.06% in <inline-formula><tex-math>$S_{alpha }$</tex-math></inline-formula>, <inline-formula><tex-math>$E_{phi }$</tex-math></inline-formula>, <inline-formula><tex-math>$F_{beta }^{omega }$</tex-math></inline-formula>, and <inline-formula><tex-math>$M$</tex-math></inline-formula>, respectively.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 2","pages":"403-415"},"PeriodicalIF":8.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148505501","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

LLM-Based Data Augmentation Method in Reinforcement Learning With Machine-Unlearning and Fine-Tuning 基于llm的基于机器学习和微调的强化学习数据增强方法

IF 8.1 3区计算机科学

IEEE Transactions on Big Data Pub Date : 2026-04-01 Epub Date: 2025-11-10 DOI: 10.1109/TBDATA.2025.3630807

Yunjiao Lei;Dayong Ye;Tianqing Zhu;Wanlei Zhou;Philip S. Yu

{"title":"LLM-Based Data Augmentation Method in Reinforcement Learning With Machine-Unlearning and Fine-Tuning","authors":"Yunjiao Lei;Dayong Ye;Tianqing Zhu;Wanlei Zhou;Philip S. Yu","doi":"10.1109/TBDATA.2025.3630807","DOIUrl":"https://doi.org/10.1109/TBDATA.2025.3630807","url":null,"abstract":"Data augmentation in reinforcement learning (RL) aims to generate diverse and extensive datasets to enhance the learning process. Most existing studies on RL augmentation employ sample-based approaches that modify existing samples. However, many of these methods directly adopt augmentation strategies from other domains, which may present limitations when applied to RL. For instance, approaches derived from computer vision techniques are often not well-suited for general RL applications that do not involve visual inputs. Moreover, such sample-based methods frequently overlook the broader characteristics of the training environment, rendering the augmented data potentially less effective in complex scenarios, such as intelligent transportation systems. In addition, these methods can introduce significant risks to critical and sensitive data. For example, introducing noise to samples as a method of augmentation can precipitate adversarial attacks. Thus, a reliable and stable method of augmentation is necessary. To address these concerns, we propose a novel large language model (LLM)-based augmentation strategy in RL with machine unlearning and fine-tuning. This method utilizes LLMs for data augmentation, ensuring the reliability of the augmented data by tailoring it to specific environmental contexts. Additionally, it enhances the quality of augmentation by mitigating the impact on critical samples through the proposed novel machine unlearning method, while simultaneously fine-tuning the model to improve overall performance. Our experimental results indicate that this innovative approach significantly surpasses traditional augmentation methods in learning performance. By mitigating the impact on critical samples, our strategy not only generates more reliable augmented data but also enhances the effectiveness of RL models.","PeriodicalId":13106,"journal":{"name":"IEEE Transactions on Big Data","volume":"12 2","pages":"536-553"},"PeriodicalIF":8.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148504388","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