Pub Date : 2024-04-01DOI: 10.1109/TETCI.2024.3375450
G. Narayanan;M. Syed Ali;Rajagopal Karthikeyan;Grienggrai Rajchakit;Sumaya Sanober;Pankaj Kumar
This paper addresses the Mittag-Leffler synchronization problem of fractional-order reaction-diffusion complex-valued neural networks (FRDCVNNs) with delays. New Mittag-Leffler synchronization (MLS) criteria in the form of the $p$ -norm for an error model derived from the drive-response model are constructed. In the design of the adaptive feedback controller, the Lyapunov approach is considered in the framework of the $p$ -norm technique, and less conservative algebraic conditions that guarantee MLS for the considered model are given. Moreover, the MLS of the considered model without reaction diffusion effect is investigated using adaptive control. Finally, an example is used to validate the proposed control scheme. To demonstrate the advantages and superiority of the proposed technique over existing methods, an image encryption method based on MLS of FRDCVNNs is considered and solved using the proposed method.
{"title":"Adaptive Strategies and its Application in the Mittag-Leffler Synchronization of Delayed Fractional-Order Complex-Valued Reaction-Diffusion Neural Networks","authors":"G. Narayanan;M. Syed Ali;Rajagopal Karthikeyan;Grienggrai Rajchakit;Sumaya Sanober;Pankaj Kumar","doi":"10.1109/TETCI.2024.3375450","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3375450","url":null,"abstract":"This paper addresses the Mittag-Leffler synchronization problem of fractional-order reaction-diffusion complex-valued neural networks (FRDCVNNs) with delays. New Mittag-Leffler synchronization (MLS) criteria in the form of the \u0000<inline-formula><tex-math>$p$</tex-math></inline-formula>\u0000-norm for an error model derived from the drive-response model are constructed. In the design of the adaptive feedback controller, the Lyapunov approach is considered in the framework of the \u0000<inline-formula><tex-math>$p$</tex-math></inline-formula>\u0000-norm technique, and less conservative algebraic conditions that guarantee MLS for the considered model are given. Moreover, the MLS of the considered model without reaction diffusion effect is investigated using adaptive control. Finally, an example is used to validate the proposed control scheme. To demonstrate the advantages and superiority of the proposed technique over existing methods, an image encryption method based on MLS of FRDCVNNs is considered and solved using the proposed method.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3294-3307"},"PeriodicalIF":5.3,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368451","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}
Pub Date : 2024-04-01DOI: 10.1109/TETCI.2024.3378599
Ling Huang;Can-Rong Guan;Zhen-Wei Huang;Yuefang Gao;Chang-Dong Wang;C. L. P. Chen
Recently, Deep Neural Networks (DNNs) have been largely utilized in Collaborative Filtering (CF) to produce more accurate recommendation results due to their ability of extracting the nonlinear relationships in the user-item pairs. However, the DNNs-based models usually encounter high computational complexity, i.e., consuming very long training time and storing huge amount of trainable parameters. To address these problems, we develop a novel broad recommender system named Broad Collaborative Filtering (BroadCF), which is an efficient nonlinear collaborative filtering approach. Instead of DNNs, Broad Learning System (BLS) is used as a mapping function to learn the nonlinear matching relationships in the user-item pairs, which can avoid the above issues while achieving very satisfactory rating prediction performance. Contrary to DNNs, BLS is a shallow network that captures nonlinear relationships between input features simply and efficiently. However, directly feeding the original rating data into BLS is not suitable due to the very large dimensionality of the original rating vector. To this end, a new preprocessing procedure is designed to generate user-item rating collaborative vector, which is a low-dimensional user-item input vector that can leverage quality judgments of the most similar users/items. Convincing experimental results on seven datasets have demonstrated the effectiveness of the BroadCF algorithm.
{"title":"Broad Recommender System: An Efficient Nonlinear Collaborative Filtering Approach","authors":"Ling Huang;Can-Rong Guan;Zhen-Wei Huang;Yuefang Gao;Chang-Dong Wang;C. L. P. Chen","doi":"10.1109/TETCI.2024.3378599","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3378599","url":null,"abstract":"Recently, Deep Neural Networks (DNNs) have been largely utilized in Collaborative Filtering (CF) to produce more accurate recommendation results due to their ability of extracting the nonlinear relationships in the user-item pairs. However, the DNNs-based models usually encounter high computational complexity, i.e., consuming very long training time and storing huge amount of trainable parameters. To address these problems, we develop a novel broad recommender system named Broad Collaborative Filtering (BroadCF), which is an efficient nonlinear collaborative filtering approach. Instead of DNNs, Broad Learning System (BLS) is used as a mapping function to learn the nonlinear matching relationships in the user-item pairs, which can avoid the above issues while achieving very satisfactory rating prediction performance. Contrary to DNNs, BLS is a shallow network that captures nonlinear relationships between input features simply and efficiently. However, directly feeding the original rating data into BLS is not suitable due to the very large dimensionality of the original rating vector. To this end, a new preprocessing procedure is designed to generate user-item rating collaborative vector, which is a low-dimensional user-item input vector that can leverage quality judgments of the most similar users/items. Convincing experimental results on seven datasets have demonstrated the effectiveness of the BroadCF algorithm.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 4","pages":"2843-2857"},"PeriodicalIF":5.3,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141965846","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}
Pub Date : 2024-04-01DOI: 10.1109/TETCI.2024.3377254
Shuai Sui;Dongyu Shen;Shaocheng Tong;C. L. Philip Chen
This study examined the problemof event-triggered formation control for nonlinear multiagent systems (MASs) with unmeasured states. First, by applying fuzzy logic systems (FLSs), the identification of unknown nonlinearities could be achieved. To save communication resources, we introduce an event-triggered mechanism. And use the triggered output signal to construct the fuzzy state observer. Then, a formation control algorithm based on event-triggered is proposed through dynamic surface control (DSC) technology and adaptive backstepping control technology, combined with two new event-triggered conditions. Finally, using the Lyapunov theory, it can be shown that all closed-loop signals are bounded. The validity of the proposed scheme can be demonstrated through simulation verification.
{"title":"State-Observer-Based Adaptive Fuzzy Event-Triggered Formation Control for Nonlinear Multiagent System","authors":"Shuai Sui;Dongyu Shen;Shaocheng Tong;C. L. Philip Chen","doi":"10.1109/TETCI.2024.3377254","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3377254","url":null,"abstract":"This study examined the problemof event-triggered formation control for nonlinear multiagent systems (MASs) with unmeasured states. First, by applying fuzzy logic systems (FLSs), the identification of unknown nonlinearities could be achieved. To save communication resources, we introduce an event-triggered mechanism. And use the triggered output signal to construct the fuzzy state observer. Then, a formation control algorithm based on event-triggered is proposed through dynamic surface control (DSC) technology and adaptive backstepping control technology, combined with two new event-triggered conditions. Finally, using the Lyapunov theory, it can be shown that all closed-loop signals are bounded. The validity of the proposed scheme can be demonstrated through simulation verification.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3327-3338"},"PeriodicalIF":5.3,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368265","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}
The robustness of neural networks in image classification is important to resist adversarial attacks. Although many researchers proposed to enhance the network robustness by inventing network training paradigms or designing network architectures, existing approaches are mainly based on a single type of networks, e.g., convolution neural networks (CNNs) or vision Transformer (ViT). Considering a recently revealed fact that CNNs and ViT can effectively defend against adversarial attacks transferred from each other, this paper aims to enhance network robustness by designing robust hybrid architecture networks containing different types of networks. To this end, we propose a hybrid architecture-based evolutionary neural architecture search approach for robust architecture design, termed HA-ENAS. Specifically, to combine or aggregate different types of networks in the same network framework, a multi-stage block-wise hybrid architecture network is first devised as the supernet, where three types of blocks (called convolution blocks, Transformer blocks, multi-layer perception blocks) are further designed as each block's candidate, and thus a hybrid architecture-based search space is established for HA-ENAS; then, the robust hybrid architecture search is formulated as an optimization problem maximizing both clean and adversarial accuracy of architectures, and an efficient multi-objective evolutionary algorithm is employed to solve the problem, where a supernet-based retraining evaluation and a surrogate model are used to mitigate coupled weight influence and reduce the whole search cost. Experimental results show that the hybrid architectures found by the proposed HA-ENAS outperform state-of-the-art single-type architectures in terms of clean accuracy and adversarial accuracy under a variety of common attacks.
在图像分类中，神经网络的鲁棒性对于抵御恶意攻击非常重要。尽管许多研究人员提出通过发明网络训练范式或设计网络架构来增强网络的鲁棒性，但现有方法主要基于单一类型的网络，如卷积神经网络（CNN）或视觉转换器（ViT）。考虑到最近揭示的一个事实，即 CNN 和 ViT 可以有效抵御相互转移的对抗性攻击，本文旨在通过设计包含不同类型网络的鲁棒混合架构网络来增强网络的鲁棒性。为此，我们提出了一种基于混合架构的鲁棒架构设计进化神经架构搜索方法，称为 HA-ENAS。具体来说，为了在同一网络框架中组合或聚合不同类型的网络，我们首先设计了一个多阶段分块式混合架构网络作为超级网络，并进一步设计了三种类型的分块（称为卷积分块、变换器分块和多层感知分块）作为每个分块的候选，从而为 HA-ENAS 建立了一个基于混合架构的搜索空间；然后，将鲁棒混合架构搜索表述为一个优化问题，使架构的清洁度和对抗精度都最大化，并采用高效的多目标进化算法来解决该问题，其中基于超网的再训练评估和代理模型用于减轻耦合权重的影响并降低整个搜索成本。实验结果表明，在各种常见攻击下，HA-ENAS 所发现的混合架构在净精度和对抗精度方面都优于最先进的单一类型架构。
{"title":"Hybrid Architecture-Based Evolutionary Robust Neural Architecture Search","authors":"Shangshang Yang;Xiangkun Sun;Ke Xu;Yuanchao Liu;Ye Tian;Xingyi Zhang","doi":"10.1109/TETCI.2024.3400867","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3400867","url":null,"abstract":"The robustness of neural networks in image classification is important to resist adversarial attacks. Although many researchers proposed to enhance the network robustness by inventing network training paradigms or designing network architectures, existing approaches are mainly based on a single type of networks, e.g., convolution neural networks (CNNs) or vision Transformer (ViT). Considering a recently revealed fact that CNNs and ViT can effectively defend against adversarial attacks transferred from each other, this paper aims to enhance network robustness by designing robust hybrid architecture networks containing different types of networks. To this end, we propose a hybrid architecture-based evolutionary neural architecture search approach for robust architecture design, termed HA-ENAS. Specifically, to combine or aggregate different types of networks in the same network framework, a multi-stage block-wise hybrid architecture network is first devised as the supernet, where three types of blocks (called convolution blocks, Transformer blocks, multi-layer perception blocks) are further designed as each block's candidate, and thus a hybrid architecture-based search space is established for HA-ENAS; then, the robust hybrid architecture search is formulated as an optimization problem maximizing both clean and adversarial accuracy of architectures, and an efficient multi-objective evolutionary algorithm is employed to solve the problem, where a supernet-based retraining evaluation and a surrogate model are used to mitigate coupled weight influence and reduce the whole search cost. Experimental results show that the hybrid architectures found by the proposed HA-ENAS outperform state-of-the-art single-type architectures in terms of clean accuracy and adversarial accuracy under a variety of common attacks.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 4","pages":"2919-2934"},"PeriodicalIF":5.3,"publicationDate":"2024-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141965868","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}
Pub Date : 2024-03-29DOI: 10.1109/TETCI.2024.3375455
Syed Thouheed Ahmed;Vinoth Kumar Venkatesan;Mahesh T R;Roopashree S;Muthukumaran Venkatesan
The global pandemic of COVID-19 has had profound and devastating effects on human life since its emergence in 2019. This viral infection predominantly impacts the respiratory system, causing a range of severity in alveolar overlapping that results in breathlessness and fatality. A novel methodology was assessed using the primary COVID-19 dataset from Kaggle, employing a federated learning ecosystem with multi-user datasets. This technique involves extracting data logs from various user repositories and datasets within the federated learning framework. Subsequently, a validation process is conducted, followed by computation utilizing a deep feature set categorization technique augmented by artificial intelligence. This augmented intelligence is showcased in a multi-layer image classification system designed for feature recognition and extraction. The training dataset, comprising 1056 data samples, is split into 647 for training and 409 for testing. Experimental outcomes highlighted a more comprehensive mapping and prioritization of features relative to attribute values. Remarkably, the proposed classification technique surpasses existing methods in accurately labeling COVID-19 detection as opposed to pneumonia and normal lung conditions in MRI/CT images.
{"title":"Augmented Intelligence Based COVID-19 Diagnostics and Deep Feature Categorization Based on Federated Learning","authors":"Syed Thouheed Ahmed;Vinoth Kumar Venkatesan;Mahesh T R;Roopashree S;Muthukumaran Venkatesan","doi":"10.1109/TETCI.2024.3375455","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3375455","url":null,"abstract":"The global pandemic of COVID-19 has had profound and devastating effects on human life since its emergence in 2019. This viral infection predominantly impacts the respiratory system, causing a range of severity in alveolar overlapping that results in breathlessness and fatality. A novel methodology was assessed using the primary COVID-19 dataset from Kaggle, employing a federated learning ecosystem with multi-user datasets. This technique involves extracting data logs from various user repositories and datasets within the federated learning framework. Subsequently, a validation process is conducted, followed by computation utilizing a deep feature set categorization technique augmented by artificial intelligence. This augmented intelligence is showcased in a multi-layer image classification system designed for feature recognition and extraction. The training dataset, comprising 1056 data samples, is split into 647 for training and 409 for testing. Experimental outcomes highlighted a more comprehensive mapping and prioritization of features relative to attribute values. Remarkably, the proposed classification technique surpasses existing methods in accurately labeling COVID-19 detection as opposed to pneumonia and normal lung conditions in MRI/CT images.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3308-3315"},"PeriodicalIF":5.3,"publicationDate":"2024-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368438","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}
Pub Date : 2024-03-29DOI: 10.1109/TETCI.2024.3378652
Keke Tang;Tianrui Lou;Weilong Peng;Nenglun Chen;Yawen Shi;Wenping Wang
Adversarial training (AT) is one of the most effective ways against adversarial attacks. However, multi-step AT is time-consuming while single-step AT is ineffective. In this paper, we propose an Energy-AT framework to make single-step AT as effective as multi-step ones, by exploiting the two properties of energy-based models (EBM). First, we utilize the Helmholtz free energy in EBM to push generated examples to be outside of the distribution boundaries of their categories, such that they are more adversarial. Second, we apply an adaptive temperature scheme in EBM to amplify the training gradients of weak adversarial examples targetedly, such that those originally hard-to-learn examples contribute to the robustification of models also. Extensive experiments validate that Energy-AT improves the robustness of models significantly to adversarial attacks in both white-box and black-box settings, and outperforms the state-of-the-art methods.
对抗性训练（AT）是对抗对抗性攻击最有效的方法之一。然而，多步骤对抗训练耗时长，单步骤对抗训练效果差。在本文中，我们利用基于能量的模型（EBM）的两个特性，提出了一个能量-AT 框架，使单步 AT 与多步 AT 一样有效。首先，我们利用 EBM 中的赫尔姆霍兹自由能，将生成的示例推到其类别分布边界之外，使其更具对抗性。其次，我们在 EBM 中应用自适应温度方案，有针对性地放大弱对抗性示例的训练梯度，从而使这些原本难以学习的示例也有助于模型的稳健化。广泛的实验验证了 Energy-AT 在白盒和黑盒环境下都能显著提高模型对对抗性攻击的鲁棒性，并且优于最先进的方法。
{"title":"Effective Single-Step Adversarial Training With Energy-Based Models","authors":"Keke Tang;Tianrui Lou;Weilong Peng;Nenglun Chen;Yawen Shi;Wenping Wang","doi":"10.1109/TETCI.2024.3378652","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3378652","url":null,"abstract":"Adversarial training (AT) is one of the most effective ways against adversarial attacks. However, multi-step AT is time-consuming while single-step AT is ineffective. In this paper, we propose an Energy-AT framework to make single-step AT as effective as multi-step ones, by exploiting the two properties of energy-based models (EBM). First, we utilize the Helmholtz free energy in EBM to push generated examples to be outside of the distribution boundaries of their categories, such that they are more adversarial. Second, we apply an adaptive temperature scheme in EBM to amplify the training gradients of weak adversarial examples targetedly, such that those originally hard-to-learn examples contribute to the robustification of models also. Extensive experiments validate that Energy-AT improves the robustness of models significantly to adversarial attacks in both white-box and black-box settings, and outperforms the state-of-the-art methods.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3396-3407"},"PeriodicalIF":5.3,"publicationDate":"2024-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368439","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}
Pub Date : 2024-03-28DOI: 10.1109/TETCI.2024.3377265
Hongsong Wen;Xing He;Tingwen Huang
In this paper, a novel projection neural network (PNN) for solving the $L_{1}$ -minimization problem is proposed, which can be applied to sparse signal reconstruction and image reconstruction. First, a one-layer PNN is designed with the projection matrix and the projection operator, which is shown to be stable in the Lyapunov sense and converges globally to the optimal solution of the $L_{1}$ -minimization problem. Then, the finite-time convergence of the proposed PNN is further investigated, with the upper bound on the convergence time given and the convergence rate analyzed. Finally, we make comparisons of our proposed PNN with the existing neural networks. Experimental results based on random Gaussian sparse signals demonstrate the effectiveness and performance of our proposed PNN. Moreover, the experiments on grayscale image reconstruction and color image reconstruction are further implemented, which sufficiently demonstrate the superiority of our proposed PNN.
{"title":"A Novel Projection Neural Network for Sparse Optimization With ${L_mathrm{{1}}}$-Minimization Problem","authors":"Hongsong Wen;Xing He;Tingwen Huang","doi":"10.1109/TETCI.2024.3377265","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3377265","url":null,"abstract":"In this paper, a novel projection neural network (PNN) for solving the \u0000<inline-formula><tex-math>$L_{1}$</tex-math></inline-formula>\u0000-minimization problem is proposed, which can be applied to sparse signal reconstruction and image reconstruction. First, a one-layer PNN is designed with the projection matrix and the projection operator, which is shown to be stable in the Lyapunov sense and converges globally to the optimal solution of the \u0000<inline-formula><tex-math>$L_{1}$</tex-math></inline-formula>\u0000-minimization problem. Then, the finite-time convergence of the proposed PNN is further investigated, with the upper bound on the convergence time given and the convergence rate analyzed. Finally, we make comparisons of our proposed PNN with the existing neural networks. Experimental results based on random Gaussian sparse signals demonstrate the effectiveness and performance of our proposed PNN. Moreover, the experiments on grayscale image reconstruction and color image reconstruction are further implemented, which sufficiently demonstrate the superiority of our proposed PNN.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3339-3351"},"PeriodicalIF":5.3,"publicationDate":"2024-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142368267","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}
Pub Date : 2024-03-28DOI: 10.1109/TETCI.2024.3402685
Zihui Luo;Chengling Jiang;Liang Liu;Xiaolong Zheng;Huadong Ma
The rapidly evolving Industrial Internet of Things (IIoT) is driving the transition from conventional manufacturing to intelligent manufacturing. Intelligent shop scheduling, as one of the essential components of intelligent manufacturing in IIoT, is desired to allocate jobs on different machines to achieve specific production targets. The flow-shop scheduling problem with batch processing machines (FSSP-BPM), which widely exists in real-world manufacturing, requires two distinct but interdependent decisions: batch formation and job scheduling. Existing approaches rely on fixed search paradigms that utilize expert knowledge to find satisfactory solutions. However, these methods struggle to ensure solution quality under real-time constraints due to the varying data distribution and the complexity of large-scale practical problems. To address this challenge, we propose a deep reinforcement learning (DRL) based method. First, we formulate the FSSP-BPM decision process as a Markov Decision Process (MDP) and design the corresponding state, action, and reward. Second, we propose a basic scheduling framework based on an encoder-decoder model with the attention mechanism. Finally, we design a batch formation module and a scheduling module trained on unlabeled multi-dimensional data. Extensive experiments on public benchmark datasets and actual production data demonstrate that the proposed method outperforms baseline algorithms and improves makespan performance by an average of 8.33%.
{"title":"Flow-Shop Scheduling Problem With Batch Processing Machines via Deep Reinforcement Learning for Industrial Internet of Things","authors":"Zihui Luo;Chengling Jiang;Liang Liu;Xiaolong Zheng;Huadong Ma","doi":"10.1109/TETCI.2024.3402685","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3402685","url":null,"abstract":"The rapidly evolving Industrial Internet of Things (IIoT) is driving the transition from conventional manufacturing to intelligent manufacturing. Intelligent shop scheduling, as one of the essential components of intelligent manufacturing in IIoT, is desired to allocate jobs on different machines to achieve specific production targets. The flow-shop scheduling problem with batch processing machines (FSSP-BPM), which widely exists in real-world manufacturing, requires two distinct but interdependent decisions: batch formation and job scheduling. Existing approaches rely on fixed search paradigms that utilize expert knowledge to find satisfactory solutions. However, these methods struggle to ensure solution quality under real-time constraints due to the varying data distribution and the complexity of large-scale practical problems. To address this challenge, we propose a deep reinforcement learning (DRL) based method. First, we formulate the FSSP-BPM decision process as a Markov Decision Process (MDP) and design the corresponding state, action, and reward. Second, we propose a basic scheduling framework based on an encoder-decoder model with the attention mechanism. Finally, we design a batch formation module and a scheduling module trained on unlabeled multi-dimensional data. Extensive experiments on public benchmark datasets and actual production data demonstrate that the proposed method outperforms baseline algorithms and improves makespan performance by an average of 8.33%.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 5","pages":"3518-3533"},"PeriodicalIF":5.3,"publicationDate":"2024-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142377012","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}
Pub Date : 2024-03-28DOI: 10.1109/TETCI.2024.3378605
Quynh Tu Ngo;Khoa Tran Phan;Abdun Mahmood;Wei Xiang
This paper considers a secure satellite downlink communication system with a hybrid intelligent reflecting surface (IRS). A robust design problem for the satellite and IRS joint beamforming is formulated to maximize the system's worst-case secrecy rate, considering practical models of the outdated channel state information and IRS power consumption. We leverage deep reinforcement learning (DRL) to solve the problem by proposing a fast DRL algorithm, namely the deep post-decision state–deterministic policy gradient (DPDS-DPG) algorithm. In DPDS-DPG, the prior known system dynamics are exploited by integrating the PDS concept into the traditional deep DPG (DDPG) algorithm, resulting in faster learning convergence. Simulation results show a faster learning convergence of 50% for DPDS-DPG compared to DDPG, with a comparable achievable system secrecy rate. Additionally, the results demonstrate system secrecy rate gains of 52% and 35% when employing active IRS and hybrid IRS, respectively, over conventional passive IRS, thereby supporting secure communications.
{"title":"Hybrid IRS-Assisted Secure Satellite Downlink Communications: A Fast Deep Reinforcement Learning Approach","authors":"Quynh Tu Ngo;Khoa Tran Phan;Abdun Mahmood;Wei Xiang","doi":"10.1109/TETCI.2024.3378605","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3378605","url":null,"abstract":"This paper considers a secure satellite downlink communication system with a hybrid intelligent reflecting surface (IRS). A robust design problem for the satellite and IRS joint beamforming is formulated to maximize the system's worst-case secrecy rate, considering practical models of the outdated channel state information and IRS power consumption. We leverage deep reinforcement learning (DRL) to solve the problem by proposing a fast DRL algorithm, namely the deep post-decision state–deterministic policy gradient (DPDS-DPG) algorithm. In DPDS-DPG, the prior known system dynamics are exploited by integrating the PDS concept into the traditional deep DPG (DDPG) algorithm, resulting in faster learning convergence. Simulation results show a faster learning convergence of 50% for DPDS-DPG compared to DDPG, with a comparable achievable system secrecy rate. Additionally, the results demonstrate system secrecy rate gains of 52% and 35% when employing active IRS and hybrid IRS, respectively, over conventional passive IRS, thereby supporting secure communications.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 4","pages":"2858-2869"},"PeriodicalIF":5.3,"publicationDate":"2024-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141965849","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}
Pub Date : 2024-03-27DOI: 10.1109/TETCI.2024.3369319
Zhiyuan Yang;Qingfu Zhang
2D GANs have yielded impressive results especially in image synthesis. However, they often encounter challenges with multi-view inconsistency due to the absence of 3D perception in their generation process. To overcome this shortcoming, 3D-aware GANs have been proposed to take advantage of both 3D representation methods, GANs, but it is very difficult to edit semantic attributes. To explore the semantic disentanglement in the 3D-aware latent space, this paper proposes a general framework, presents two representative approaches for the 3D manipulation task in both supervised, unsupervised manners. Our key idea is to utilize existing latent discovery methods, bring direct compatibility to 3D control. Specifically, we propose a novel module to extract the semantic latent space of the existing 3D-aware models, then develop two approaches to find a normal editing direction in the latent space. Leveraging the meaningful semantic latent directions, we can easily edit the shape, appearance attributes while preserving the 3D consistency. Quantitative, qualitative experiments show that our method is effective, efficient for the 3D-aware generation with steerability on both synthetic, real-world datasets.
二维 GAN 取得了令人瞩目的成果，尤其是在图像合成方面。然而，由于在生成过程中缺乏三维感知，它们经常会遇到多视图不一致的难题。为了克服这一缺陷，人们提出了三维感知 GAN，以利用三维表示方法和 GAN 的优势，但编辑语义属性非常困难。为了探索三维感知潜空间中的语义分解问题，本文提出了一个总体框架，并针对三维操作任务提出了两种有监督和无监督的代表性方法。我们的主要想法是利用现有的潜在发现方法，直接兼容三维控制。具体来说，我们提出了一个新模块来提取现有三维感知模型的语义潜空间，然后开发了两种方法来寻找潜空间中的法线编辑方向。利用有意义的语义潜在方向，我们可以轻松地编辑形状、外观属性，同时保持三维一致性。定量和定性实验表明，我们的方法在合成和真实世界数据集上都能有效、高效地生成具有可转向性的三维感知模型。
{"title":"Discovering Interpretable Latent Space Directions for 3D-Aware Image Generation","authors":"Zhiyuan Yang;Qingfu Zhang","doi":"10.1109/TETCI.2024.3369319","DOIUrl":"https://doi.org/10.1109/TETCI.2024.3369319","url":null,"abstract":"2D GANs have yielded impressive results especially in image synthesis. However, they often encounter challenges with multi-view inconsistency due to the absence of 3D perception in their generation process. To overcome this shortcoming, 3D-aware GANs have been proposed to take advantage of both 3D representation methods, GANs, but it is very difficult to edit semantic attributes. To explore the semantic disentanglement in the 3D-aware latent space, this paper proposes a general framework, presents two representative approaches for the 3D manipulation task in both supervised, unsupervised manners. Our key idea is to utilize existing latent discovery methods, bring direct compatibility to 3D control. Specifically, we propose a novel module to extract the semantic latent space of the existing 3D-aware models, then develop two approaches to find a normal editing direction in the latent space. Leveraging the meaningful semantic latent directions, we can easily edit the shape, appearance attributes while preserving the 3D consistency. Quantitative, qualitative experiments show that our method is effective, efficient for the 3D-aware generation with steerability on both synthetic, real-world datasets.","PeriodicalId":13135,"journal":{"name":"IEEE Transactions on Emerging Topics in Computational Intelligence","volume":"8 3","pages":"2570-2580"},"PeriodicalIF":5.3,"publicationDate":"2024-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141094885","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}
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：