Knowledge-Based Systems最新文献_第6页

ConUMIP: Continuous-time dynamic graph learning via uncertainty masked mix-up on representation space ConUMIP：通过表征空间上的不确定性掩蔽混合进行连续时间动态图学习

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-15 DOI: 10.1016/j.knosys.2024.112748

Haoyu Zhang, Xuchu Jiang

{"title":"ConUMIP: Continuous-time dynamic graph learning via uncertainty masked mix-up on representation space","authors":"Haoyu Zhang, Xuchu Jiang","doi":"10.1016/j.knosys.2024.112748","DOIUrl":"10.1016/j.knosys.2024.112748","url":null,"abstract":"<div><div>Representation learning on continuous-time dynamic graphs has garnered substantial attention for its capacity to model evolving entity relationships. However, existing methods exhibit pronounced overfitting, particularly in complex and sparse data scenarios. We empirically substantiate this overfitting through multiple indicators: (1) a significant performance discrepancy between training and validation/test sets, especially for long-term interaction predictions; (2) an inverse correlation between model complexity and generalization performance; (3) a widening temporal generalization gap as the prediction horizons extend; and (4) rapid performance deterioration under data-sparse conditions. These phenomena collectively demonstrate the overfitting issue, limiting the applicability of current approaches in cold-start scenarios and dynamic environments. To address this, we propose <strong>Con</strong>tinuous-Time Dynamic Graph Learning via <strong>U</strong>ncertainty <strong>M</strong>asked M<strong>I</strong>x-U<strong>P</strong> (ConUMIP), a novel data augmentation method operating in the representation space of continuous-time dynamic graphs. Unlike conventional techniques that perturb raw graph data, ConUMIP adaptively captures temporal evolution patterns and generates diverse augmented samples. This approach effectively mitigates overfitting while enhancing long-term dependency modeling. By eschewing predefined time windows and integrating both local and global structures, ConUMIP demonstrates superior adaptation to complex dynamic evolution patterns. Comprehensive evaluations across five real-world datasets validate ConUMIP's efficacy in substantially improving both the performance and generalizability of existing continuous-time dynamic graph models, particularly in long-term predictions and data-sparse scenarios, without incurring additional computational complexity, thus offering a robust solution to the overfitting challenge in this domain.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112748"},"PeriodicalIF":7.2,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

CSDD-Net: A cross semi-supervised dual-feature distillation network for industrial defect detection CSDD-Net：用于工业缺陷检测的交叉半监督双特征蒸馏网络

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-13 DOI: 10.1016/j.knosys.2024.112751

Mingle Zhou , Zhanzhi Su , Min Li , Yingjie Wang , Gang Li

{"title":"CSDD-Net: A cross semi-supervised dual-feature distillation network for industrial defect detection","authors":"Mingle Zhou , Zhanzhi Su , Min Li , Yingjie Wang , Gang Li","doi":"10.1016/j.knosys.2024.112751","DOIUrl":"10.1016/j.knosys.2024.112751","url":null,"abstract":"<div><div>Detecting defects in industrial products is crucial to the strict quality control of products. Most current methods focus on supervised learning, relying on large-scale labeled samples. However, the forms of defects in industrial scenarios vary, and the data collection cost is high, which makes it difficult to meet the high requirements of massive labeled data. Therefore, we propose a Cross Semi-Supervised Dual-Feature Distillation Network (CSDD-Net), which aims to cross-use supervised and semi-supervised networks to learn rich feature representations and the distribution of large-scale features, respectively. CSDD-Net can transfer the defect feature distribution learned on partially labeled data in supervised branch to unsupervised branch, achieving simultaneous modeling and distillation based on partially labeled data. Firstly, this paper proposes a cross-local-global feature extraction network. By designing double interaction and ghost linear attention structure, it aims to force the network to be able to focus on local detail texture in global features and local features to perceive global semantics. Secondly, this paper proposes a Closed-Loop Cross-Aggregation Network (CLCA-Net), which considers deep and shallow semantics and fine-grained information. Thirdly, this paper designs a dynamic adaptive distillation loss, which could automatically adjust a more suitable regression loss function according to the defect characteristics, ensuring that the model could accurately locate and regress defects of various scales. Finally, this paper proposes a Glass Bottleneck defect dataset and verifies the feasibility of CSDD-Net in practical industrial applications. CSDD-Net achieved [email protected] of 80.41%, 76.42%, and 97.12% on the Glass Bottleneck, Wood, and Aluminum datasets with only 13.5 GFLOPs.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112751"},"PeriodicalIF":7.2,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Multi-agent reinforcement learning with synchronized and decomposed reward automaton synthesized from reactive temporal logic 利用由反应时态逻辑合成的同步和分解奖励自动机进行多代理强化学习

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112703

Chenyang Zhu , Jinyu Zhu , Wen Si , Xueyuan Wang , Fang Wang

{"title":"Multi-agent reinforcement learning with synchronized and decomposed reward automaton synthesized from reactive temporal logic","authors":"Chenyang Zhu , Jinyu Zhu , Wen Si , Xueyuan Wang , Fang Wang","doi":"10.1016/j.knosys.2024.112703","DOIUrl":"10.1016/j.knosys.2024.112703","url":null,"abstract":"<div><div>Multi-agent systems (MAS) consist of multiple autonomous agents interacting to achieve collective objectives. Multi-agent reinforcement learning (MARL) enhances these systems by enabling agents to learn optimal behaviors through interaction, thus improving their coordination in dynamic environments. However, MARL faces significant challenges in adapting to complex dependencies on past states and actions, which are not adequately represented by the current state alone in reactive systems. This paper addresses these challenges by considering MAS operating under task specifications formulated as Generalized Reactivity of rank 1 (GR(1)). These synthesized strategies are used as a priori knowledge to guide the learning. To tackle the difficulties of handling non-Markovian tasks in reactive systems, we propose a novel synchronized decentralized training paradigm that guides agents to learn within the MARL framework using a reward structure constructed from decomposed synthesized strategies of GR(1). We initially formalize the synthesis of GR(1) strategies as a reachability problem of winning states of the system. Subsequently, we develop a decomposition mechanism that constructs individual reward structures for decentralized MARL, incorporating potential values calculated through value iteration. Theoretical proofs are provided to verify that the safety and liveness properties are preserved. We evaluate our approach against other state-of-the-art methods under various GR(1) specifications and scenario maps, demonstrating superior learning efficacy and optimal rewards per episode. Additionally, we show that the decentralized training paradigm outperforms the centralized training paradigm. The value iteration strategy used to calculate potential values for the reward structure is compared against two other strategies, showcasing its advantages.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112703"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142705015","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

CESDQL: Communicative experience-sharing deep Q-learning for scalability in multi-robot collaboration with sparse reward CESDQL：在奖励稀疏的多机器人协作中利用交流经验共享深度 Q-learning 实现可扩展性

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112714

Muhammad Naveed Abbas , Paul Liston , Brian Lee , Yuansong Qiao

{"title":"CESDQL: Communicative experience-sharing deep Q-learning for scalability in multi-robot collaboration with sparse reward","authors":"Muhammad Naveed Abbas , Paul Liston , Brian Lee , Yuansong Qiao","doi":"10.1016/j.knosys.2024.112714","DOIUrl":"10.1016/j.knosys.2024.112714","url":null,"abstract":"<div><div>Owing to the massive transformation in industrial processes and logistics, warehouses are also undergoing advanced automation. The application of Autonomous Mobile Robots (a.k.a. multi-robots) is one of the important elements of overall warehousing automation. The autonomous collaborative behaviour of the multi-robots can be considered as employment on a control task and, thus, can be optimised using multi-agent reinforcement learning (MARL). Consequently, an autonomous warehouse is to be represented by an MARL environment. An MARL environment replicating an autonomous warehouse poses the challenge of exploration due to sparse reward leading to inefficient collaboration. This challenge aggravates further with an increase in the number of robots and the grid size, i.e., scalability. This research proposes <strong>C</strong>ommunicative <strong>E</strong>xperience-<strong>S</strong>haring <strong>D</strong>eep <strong>Q</strong>-<strong>L</strong>earning (CESDQL) based on Q-learning, a novel hybrid multi-robot communicative framework for scalability for MARL collaboration with sparse rewards, where exploration is challenging and makes collaboration difficult. CESDQL makes use of experience-sharing through collective sampling from the Experience (Replay) buffer and communication through Communicative Deep recurrent Q-network (CommDRQN), a Q-function approximator. Through empirical evaluation of CESDQL in a variety of collaborative scenarios, it is established that CESDQL outperforms the baselines in terms of convergence and stable learning. Overall, CESDQL achieves 5%, 69%, 60%, 211%, 171%, 3.8% & 10% more final accumulative training returns than the closest performing baseline by scenario, and, 27%, 10.33% & 573% more final average training returns than the closest performing baseline by the big-scale scenario.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112714"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142705006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Boosting point cloud understanding through graph convolutional network with scale measurement and high-frequency enhancement 通过带有比例测量和高频增强功能的图卷积网络提升点云理解能力

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112715

Yun Bai , Guanlin Li , Xuchao Gong , Kuijie Zhang , Qian Xiao , Chaozhi Yang , Zongmin Li

{"title":"Boosting point cloud understanding through graph convolutional network with scale measurement and high-frequency enhancement","authors":"Yun Bai , Guanlin Li , Xuchao Gong , Kuijie Zhang , Qian Xiao , Chaozhi Yang , Zongmin Li","doi":"10.1016/j.knosys.2024.112715","DOIUrl":"10.1016/j.knosys.2024.112715","url":null,"abstract":"<div><div>Graph-based methods have exhibited exceptional performance in point cloud understanding by capturing local geometric relationships. However, existing approaches often struggle to characterize the overall spatial scale of local graphs. In addition, they fail to capture the differences between nodes effectively, which is crucial for distinguishing different classes. This study introduces SM-HFEGCN, a novel graph convolutional network that addresses these limitations through two key innovations: scale measurement and high-frequency enhancement. First, we introduce a spatial scale feature derived from the diagonal vectors of the neighborhood, which serves as a unique graph-specific property related to the geometry and density of the local point cloud. This feature can characterize the overall spatial scale of the local point cloud. Second, we enhance the high-frequency information to capture node variations and integrate it with smoothed information to represent the differences and similarities between nodes simultaneously. Extensive experiments demonstrate the effectiveness of SM-HFEGCN in point cloud classification and segmentation tasks.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112715"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

FourierAugment: Frequency-based image encoding for resource-constrained vision tasks 傅立叶增强：基于频率的图像编码，用于资源受限的视觉任务

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112695

Jiae Yoon , Myeongjin Lee , Ue-Hwan Kim

引用次数: 0

Bridging the gap between text-to-SQL research and real-world applications: A unified all-in-one framework for text-to-SQL 缩小文本到 SQL 研究与实际应用之间的差距：统一的文本到 SQL 一体化框架

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112697

Mirae Han , Seongsik Park , Seulgi Kim , Harksoo Kim

引用次数: 0

Advanced approach for mining utility occupancy patterns in incremental environment 在增量环境中挖掘公用设施占用模式的先进方法

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-12 DOI: 10.1016/j.knosys.2024.112713

Myungha Cho , Hanju Kim , Seungwan Park , Doyoung Kim, Doyoon Kim, Unil Yun

{"title":"Advanced approach for mining utility occupancy patterns in incremental environment","authors":"Myungha Cho , Hanju Kim , Seungwan Park , Doyoung Kim, Doyoon Kim, Unil Yun","doi":"10.1016/j.knosys.2024.112713","DOIUrl":"10.1016/j.knosys.2024.112713","url":null,"abstract":"<div><div>In recent years, one of the varied fields of mining techniques that can discover valuable patterns from databases with vast amounts of data, utility pattern mining, has been studied. Besides, pattern mining techniques considering utility occupancy have been developed, considering the profit, quantity, and proportion of the pattern in transactions. However, recent pattern mining studies for utility occupancy still suffer from obtaining patterns in an incremental environment. Meanwhile, with the widespread adoption of technologies such as IoT or networks, data is rapidly generated and accumulated between devices in real time. Therefore, we suggest IUOIL (Incremental high-Utility Occupancy pattern mining with Indexed List) that discovers patterns having high utility occupancy employing an indexed list-based data structure from databases in an incremental environment. Our algorithm can obtain results by quickening the combination process for patterns using the data structure and reducing search space with three efficient pruning strategies. Performance evaluation is performed using various datasets for comparison with existing algorithms. The assessment on real datasets demonstrated that the technique extracts exact results with the fastest runtime while minimizing memory consumption. In addition, the evaluations on synthetic datasets showed that the technique discovers result set of patterns efficiently and stably as the volume of a database increases.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112713"},"PeriodicalIF":7.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142705187","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

DRN-DSA: A hybrid deep learning network model for precipitation nowcasting using time series data DRN-DSA：利用时间序列数据进行降水预报的混合深度学习网络模型

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-10 DOI: 10.1016/j.knosys.2024.112679

Gujanatti Rudrappa , Nataraj Vijapur

{"title":"DRN-DSA: A hybrid deep learning network model for precipitation nowcasting using time series data","authors":"Gujanatti Rudrappa , Nataraj Vijapur","doi":"10.1016/j.knosys.2024.112679","DOIUrl":"10.1016/j.knosys.2024.112679","url":null,"abstract":"<div><div>Precipitation nowcasting involves short-term weather forecasting, predicting rain or snow within the next two hours. By analyzing current atmospheric conditions, it aids meteorologists in identifying weather patterns and preparing for severe events such as flooding. These nowcasts are typically displayed on geographical maps by weather services. However, the rapidly changing climate conditions make precipitation nowcasting a formidable challenge, as accurate short-term forecasts are hindered by immediate weather fluctuations. Traditional nowcasting methods, like numerical models and radar extrapolation, have limitations in delivering highly detailed and timely precipitation nowcasts. To overcome this issue, an effective solution is framed for precipitation nowcasting using a hybrid network approach named Deep Residual Network-Deep Stacked Autoencoder (DRN-DSA). Initially, the input time series data is acquired from the dataset. Thereafter, the effective technical indicators are extracted at the feature extraction stage. Later on, precipitation-type nowcasting is carried out using the proposed hybrid DRN-DSA, which is developed by incorporating a Deep Stacked Autoencoder (DSA) and Deep Residual Network (DRN). Finally, Weather nowcasting is carried out using the same proposed hybrid DSA-DRN. Moreover, when compared to other traditional models, the proposed DRN-DSA has gained superior results with a Relative Absolute Error (RAE) of 0.295, Root Mean Square Error (RMSE) of 0.154, low Mean Square Error (MSE) of 0.0236, Mean Absolute Percentage Error (MAPE) of 0.295, and False Acceptance Rate (FAR) of 0.0118.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112679"},"PeriodicalIF":7.2,"publicationDate":"2024-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142706016","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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

Fitness and historical success information-assisted binary particle swarm optimization for feature selection 用于特征选择的适合度和历史成功信息辅助二元粒子群优化法

IF 7.2 1区计算机科学

Knowledge-Based Systems Pub Date : 2024-11-10 DOI: 10.1016/j.knosys.2024.112699

Shubham Gupta, Saurabh Gupta

{"title":"Fitness and historical success information-assisted binary particle swarm optimization for feature selection","authors":"Shubham Gupta, Saurabh Gupta","doi":"10.1016/j.knosys.2024.112699","DOIUrl":"10.1016/j.knosys.2024.112699","url":null,"abstract":"<div><div>Feature selection is a critical preprocessing step in machine learning aimed at identifying the most relevant features or variables from a dataset. Although conventional particle swarm optimization (PSO) has shown efficiency for feature selection tasks, developing an effective PSO algorithm for this task is still challenging. This study proposes a fitness and historical success information-assisted binary particle swarm optimization, denoted by FPSO. The FPSO is developed by integrating different search strategies, including a weighted center-based approach, historical information-based acceleration coefficients, and selection operation. These strategies are embedded into the FPSO to enhance the levels of exploration and exploitation based on the fitness value of particles and their historical search status. In the FPSO, the transfer function is also added to transform the continuous search space into binary search space. Experimental validation and comparison with seven other metaheuristic algorithms on 24 datasets verify the effectiveness of the FPSO in eliminating irrelevant and redundant features.</div></div>","PeriodicalId":49939,"journal":{"name":"Knowledge-Based Systems","volume":"306 ","pages":"Article 112699"},"PeriodicalIF":7.2,"publicationDate":"2024-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142705190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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