Due to its strong ability in distribution representation, the diffusion model has been incorporated into offline reinforcement learning (RL) to cover diverse trajectories of the complex behavior policy. However, this also causes several challenges. Training the diffusion model to imitate behavior from the collected trajectories suffers from limited stitching capability which derives better policies from suboptimal trajectories. Furthermore, the inherent randomness of the diffusion model can lead to unpredictable control and dangerous behavior for the robot. To address these concerns, we propose the value-learning-based decision diffuser (V-DD), which consists of the trajectory diffusion module (TDM) and the trajectory evaluation module (TEM). During the training process, the TDM combines the state-value and classifier-free guidance to bolster the ability to stitch suboptimal trajectories. During the inference process, we design the TEM to select a feasible trajectory generated by the diffusion model. Empirical results demonstrate that our method delivers competitive results on the D4RL benchmark and substantially outperforms current diffusion model-based methods on the real-world robot task.
由于扩散模型在分布表示方面的强大能力，它已被纳入离线强化学习（RL），以覆盖复杂行为政策的各种轨迹。然而，这也带来了一些挑战。从收集到的轨迹中训练扩散模型来模仿行为，会受到拼接能力的限制，从而从次优轨迹中得出更好的策略。此外，扩散模型固有的随机性可能会导致机器人无法预测的控制和危险行为。为了解决这些问题，我们提出了基于价值学习的决策扩散器（V-DD），它由轨迹扩散模块（TDM）和轨迹评估模块（TEM）组成。在训练过程中，TDM 结合了状态值和无分类器指导，以提高缝合次优轨迹的能力。在推理过程中，我们设计 TEM 来选择由扩散模型生成的可行轨迹。实证结果表明，我们的方法在 D4RL 基准测试中取得了具有竞争力的结果，并且在实际机器人任务中大大优于当前基于扩散模型的方法。
{"title":"Stabilizing Diffusion Model for Robotic Control With Dynamic Programming and Transition Feasibility","authors":"Haoran Li;Yaocheng Zhang;Haowei Wen;Yuanheng Zhu;Dongbin Zhao","doi":"10.1109/TAI.2024.3387401","DOIUrl":"https://doi.org/10.1109/TAI.2024.3387401","url":null,"abstract":"Due to its strong ability in distribution representation, the diffusion model has been incorporated into offline reinforcement learning (RL) to cover diverse trajectories of the complex behavior policy. However, this also causes several challenges. Training the diffusion model to imitate behavior from the collected trajectories suffers from limited stitching capability which derives better policies from suboptimal trajectories. Furthermore, the inherent randomness of the diffusion model can lead to unpredictable control and dangerous behavior for the robot. To address these concerns, we propose the value-learning-based decision diffuser (V-DD), which consists of the trajectory diffusion module (TDM) and the trajectory evaluation module (TEM). During the training process, the TDM combines the state-value and classifier-free guidance to bolster the ability to stitch suboptimal trajectories. During the inference process, we design the TEM to select a feasible trajectory generated by the diffusion model. Empirical results demonstrate that our method delivers competitive results on the D4RL benchmark and substantially outperforms current diffusion model-based methods on the real-world robot task.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-10DOI: 10.1109/TAI.2024.3387406
Kangjia He;Li Liu;Youmin Zhang;Ye Wang;Qun Liu;Guoyin Wang
Counterfactual subgraphs explain graph neural networks (GNNs) by answering the question: “How would the prediction change if a certain subgraph were absent in the input instance?” The differentiable proxy adjacency matrix is prevalent in current counterfactual subgraph discovery studies due to its ability to avoid exhaustive edge searching. However, a prediction gap exists when feeding the proxy matrix with continuous values and the thresholded discrete adjacency matrix to GNNs, compromising the optimization of the subgraph generator. Furthermore, the end-to-end learning schema adopted in the subgraph generator limits the diversity of counterfactual subgraphs. To this end, we propose CF-GFNExplainer, a flow-based approach for learning counterfactual subgraphs. CF-GFNExplainer employs a policy network with a discrete edge removal schema to construct counterfactual subgraph generation trajectories. Additionally, we introduce a loss function designed to guide CF-GFNExplainer's optimization. The discrete adjacency matrix generated in each trajectory eliminates the prediction gap, enhancing the validity of the learned subgraphs. Furthermore, the multitrajectories sampling strategy adopted in CF-GFNExplainer results in diverse counterfactual subgraphs. Extensive experiments conducted on synthetic and real-world datasets demonstrate the effectiveness of the proposed method in terms of validity and diversity.
{"title":"Learning Counterfactual Explanation of Graph Neural Networks via Generative Flow Network","authors":"Kangjia He;Li Liu;Youmin Zhang;Ye Wang;Qun Liu;Guoyin Wang","doi":"10.1109/TAI.2024.3387406","DOIUrl":"https://doi.org/10.1109/TAI.2024.3387406","url":null,"abstract":"Counterfactual subgraphs explain graph neural networks (GNNs) by answering the question: “How would the prediction change if a certain subgraph were absent in the input instance?” The differentiable proxy adjacency matrix is prevalent in current counterfactual subgraph discovery studies due to its ability to avoid exhaustive edge searching. However, a prediction gap exists when feeding the proxy matrix with continuous values and the thresholded discrete adjacency matrix to GNNs, compromising the optimization of the subgraph generator. Furthermore, the end-to-end learning schema adopted in the subgraph generator limits the diversity of counterfactual subgraphs. To this end, we propose CF-GFNExplainer, a flow-based approach for learning counterfactual subgraphs. CF-GFNExplainer employs a policy network with a discrete edge removal schema to construct counterfactual subgraph generation trajectories. Additionally, we introduce a loss function designed to guide CF-GFNExplainer's optimization. The discrete adjacency matrix generated in each trajectory eliminates the prediction gap, enhancing the validity of the learned subgraphs. Furthermore, the multitrajectories sampling strategy adopted in CF-GFNExplainer results in diverse counterfactual subgraphs. Extensive experiments conducted on synthetic and real-world datasets demonstrate the effectiveness of the proposed method in terms of validity and diversity.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-10DOI: 10.1109/TAI.2024.3386499
Ziyan Zhang;Bo Jiang;Jin Tang;Jinhui Tang;Bin Luo
Graph convolutional networks (GCNs) gain increasing attention on graph data learning tasks in recent years. However, in many applications, graph may come with an incomplete form where attributes of graph nodes are partially unknown/missing. Existing graph convolutions (GCs) are generally designed on complete graphs which cannot deal with attribute-incomplete graph data directly. To address this problem, in this article, we extend standard GC and develop an explicit Partial Graph Convolution (PaGC) for attribute-incomplete graph data. Our PaGC is derived based on the observation that the core neighborhood aggregator in GC operation can be equivalently viewed as an energy minimization model. Based on it, we can define a novel partial aggregation function and derive PaGC for incomplete graph data. Experiments demonstrate the effectiveness and efficiency of the proposed PaGCN.
{"title":"Incomplete Graph Learning via Partial Graph Convolutional Network","authors":"Ziyan Zhang;Bo Jiang;Jin Tang;Jinhui Tang;Bin Luo","doi":"10.1109/TAI.2024.3386499","DOIUrl":"https://doi.org/10.1109/TAI.2024.3386499","url":null,"abstract":"Graph convolutional networks (GCNs) gain increasing attention on graph data learning tasks in recent years. However, in many applications, graph may come with an incomplete form where attributes of graph nodes are partially unknown/missing. Existing graph convolutions (GCs) are generally designed on complete graphs which cannot deal with attribute-incomplete graph data directly. To address this problem, in this article, we extend standard GC and develop an explicit Partial Graph Convolution (PaGC) for attribute-incomplete graph data. Our PaGC is derived based on the observation that the core neighborhood aggregator in GC operation can be equivalently viewed as an energy minimization model. Based on it, we can define a novel \u0000<italic>partial aggregation function</i>\u0000 and derive PaGC for incomplete graph data. Experiments demonstrate the effectiveness and efficiency of the proposed PaGCN.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-09DOI: 10.1109/TAI.2024.3386500
Arunava Roy;Dipankar Dasgupta
Generative adversarial networks (GANs) excel in diverse applications such as image enhancement, manipulation, and generating images and videos from text. Yet, training GANs with large datasets remains computationally intensive for standalone systems. Synchronization issues between the generator and discriminator lead to unstable training, poor convergence, vanishing, and exploding gradient challenges. In decentralized environments, standalone GANs struggle with distributed data on client machines. Researchers have turned to federated learning (FL) for distributed-GAN (D-GAN) implementations, but efforts often fall short due to training instability and poor synchronization within GAN components. In this study, we present DRL-GAN, a lightweight Wasserstein conditional distributed relativistic loss-GAN designed to overcome existing limitations. DRL-GAN ensures training stability in the face of nonconvex losses by employing a single global generator on the central server and a discriminator per client. Utilizing Wasserstein-1 for relativistic loss computation between real and fake samples, DRL-GAN effectively addresses issues, such as mode collapses, vanishing, and exploding gradients, accommodating both iid and non-iid private data in clients and fostering strong convergence. The absence of a robust conditional distributed-GAN model serves as another motivation for this work. We provide a comprehensive mathematical formulation of DRL-GAN and validate our claims empirically on CIFAR-10, MNIST, EuroSAT, and LSUN-Bedroom datasets.
{"title":"A Distributed Conditional Wasserstein Deep Convolutional Relativistic Loss Generative Adversarial Network With Improved Convergence","authors":"Arunava Roy;Dipankar Dasgupta","doi":"10.1109/TAI.2024.3386500","DOIUrl":"https://doi.org/10.1109/TAI.2024.3386500","url":null,"abstract":"Generative adversarial networks (GANs) excel in diverse applications such as image enhancement, manipulation, and generating images and videos from text. Yet, training GANs with large datasets remains computationally intensive for standalone systems. Synchronization issues between the generator and discriminator lead to unstable training, poor convergence, vanishing, and exploding gradient challenges. In decentralized environments, standalone GANs struggle with distributed data on client machines. Researchers have turned to federated learning (FL) for distributed-GAN (D-GAN) implementations, but efforts often fall short due to training instability and poor synchronization within GAN components. In this study, we present DRL-GAN, a lightweight Wasserstein conditional distributed relativistic loss-GAN designed to overcome existing limitations. DRL-GAN ensures training stability in the face of nonconvex losses by employing a single global generator on the central server and a discriminator per client. Utilizing Wasserstein-1 for relativistic loss computation between real and fake samples, DRL-GAN effectively addresses issues, such as mode collapses, vanishing, and exploding gradients, accommodating both iid and non-iid private data in clients and fostering strong convergence. The absence of a robust conditional distributed-GAN model serves as another motivation for this work. We provide a comprehensive mathematical formulation of DRL-GAN and validate our claims empirically on CIFAR-10, MNIST, EuroSAT, and LSUN-Bedroom datasets.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-08DOI: 10.1109/TAI.2024.3386149
Handuo Yang;Ju Huyan;Tao Ma;Yitao Song;Chengjia Han
To address two key challenges—limited grid-level detection capability and difficulty in detecting pavement cracks in complex environments, this study proposes a novel neural network model called CrackcellNet. This innovative model incorporates an output structure that enables end-to-end grid recognition and a module that enhances shadow image data to enhance crack detection. The model relies on the design of consecutive pooling layers to achieve adaptive target size grid output. By utilizing image fusion techniques, it enhances the quantity of shadow data in road surface detection. The results of ablation experiments indicate that the optimal configuration for CrackcellNet includes V-block and shadow augmentation operations, dilation rates of 1 or 2, and a convolutional layer in the CBA module. Through extensive experimentation, we have demonstrated that our model achieved an accuracy rate of 94.5% for grid-level crack detection and a F1 value of 0.839. Furthermore, practical engineering validation confirms the model's efficacy with an average PCIe of 0.045, providing valuable guidance for road maintenance decisions.
为了解决网格级检测能力有限和复杂环境下路面裂缝检测困难这两大难题，本研究提出了一种名为 CrackcellNet 的新型神经网络模型。这一创新模型包含一个可实现端到端网格识别的输出结构和一个可增强阴影图像数据以提高裂缝检测能力的模块。该模型依靠连续池化层的设计来实现自适应目标尺寸网格输出。通过利用图像融合技术，该模型增强了路面检测中阴影数据的数量。烧蚀实验结果表明，CrackcellNet 的最佳配置包括 V 块和阴影增强操作、1 或 2 的扩张率以及 CBA 模块中的卷积层。通过大量实验，我们证明了我们的模型在网格级裂纹检测方面达到了 94.5% 的准确率和 0.839 的 F1 值。此外，实际工程验证也证实了该模型的有效性，其平均 PCIe 为 0.045，为道路维护决策提供了宝贵的指导。
{"title":"A Novel Applicable Shadow Resistant Neural Network Model for High-Efficiency Grid-Level Pavement Crack Detection","authors":"Handuo Yang;Ju Huyan;Tao Ma;Yitao Song;Chengjia Han","doi":"10.1109/TAI.2024.3386149","DOIUrl":"https://doi.org/10.1109/TAI.2024.3386149","url":null,"abstract":"To address two key challenges—limited grid-level detection capability and difficulty in detecting pavement cracks in complex environments, this study proposes a novel neural network model called CrackcellNet. This innovative model incorporates an output structure that enables end-to-end grid recognition and a module that enhances shadow image data to enhance crack detection. The model relies on the design of consecutive pooling layers to achieve adaptive target size grid output. By utilizing image fusion techniques, it enhances the quantity of shadow data in road surface detection. The results of ablation experiments indicate that the optimal configuration for CrackcellNet includes V-block and shadow augmentation operations, dilation rates of 1 or 2, and a convolutional layer in the CBA module. Through extensive experimentation, we have demonstrated that our model achieved an accuracy rate of 94.5% for grid-level crack detection and a F1 value of 0.839. Furthermore, practical engineering validation confirms the model's efficacy with an average PCIe of 0.045, providing valuable guidance for road maintenance decisions.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165065","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-05DOI: 10.1109/TAI.2024.3385384
Guifeng Ye;Shaowen Lu
Our proposed algorithm is a prefetching-based multiproposal Markov Chain Monte Carlo (PMP-MCMC) method that efficiently explores the target distribution by combining multiple proposals with the concept of prefetching. In our method, not all proposals are directly derived from the current state; some are derived from future states. This approach breaks through the inherent sequential characteristics of traditional MCMC algorithms. Compared with single-proposal and multiproposal methods, our approach speeds up by $K$ times and the burn-in period is reduced by a factor of $1/text{log}_{2}K$ maximally, where $K$ is the number of parallel computational units or processing cores. Compared with prefetching method, our method has increased the number of samples per iteration by a factor of $K/text{log}_{2}K$ . Furthermore, the proposed method is general and can be integrated into MCMC variants such as Hamiltonian Monte Carlo (HMC). We have also applied this method to optimize the model parameters of neural networks and Bayesian inference and observed significant improvements in optimization performance.
{"title":"Prefetching-based Multiproposal Markov Chain Monte Carlo Algorithm","authors":"Guifeng Ye;Shaowen Lu","doi":"10.1109/TAI.2024.3385384","DOIUrl":"https://doi.org/10.1109/TAI.2024.3385384","url":null,"abstract":"Our proposed algorithm is a prefetching-based multiproposal Markov Chain Monte Carlo (PMP-MCMC) method that efficiently explores the target distribution by combining multiple proposals with the concept of prefetching. In our method, not all proposals are directly derived from the current state; some are derived from future states. This approach breaks through the inherent sequential characteristics of traditional MCMC algorithms. Compared with single-proposal and multiproposal methods, our approach speeds up by \u0000<inline-formula><tex-math>$K$</tex-math></inline-formula>\u0000 times and the burn-in period is reduced by a factor of \u0000<inline-formula><tex-math>$1/text{log}_{2}K$</tex-math></inline-formula>\u0000 maximally, where \u0000<inline-formula><tex-math>$K$</tex-math></inline-formula>\u0000 is the number of parallel computational units or processing cores. Compared with prefetching method, our method has increased the number of samples per iteration by a factor of \u0000<inline-formula><tex-math>$K/text{log}_{2}K$</tex-math></inline-formula>\u0000. Furthermore, the proposed method is general and can be integrated into MCMC variants such as Hamiltonian Monte Carlo (HMC). We have also applied this method to optimize the model parameters of neural networks and Bayesian inference and observed significant improvements in optimization performance.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165064","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-05DOI: 10.1109/TAI.2024.3385390
Tiehang Duan;Zhenyi Wang;Li Shen;Gianfranco Doretto;Donald A. Adjeroh;Fang Li;Cui Tao
Large variance exists in Electroencephalogram (EEG) signals with its pattern differing significantly across subjects. It is a challenging problem to perform online sequential decoding of EEG signals across different subjects, where a sequence of subjects arrive in temporal order and no signal data is jointly available beforehand. The challenges include the following two aspects: 1) the knowledge learned from previous subjects does not readily fit to future subjects, and fast adaptation is needed in the process; and 2) the EEG classifier could drastically erase information of learnt subjects as learning progresses, namely catastrophic forgetting. Most existing EEG decoding explorations use sizable data for pretraining purposes, and to the best of our knowledge we are the first to tackle this challenging online sequential decoding setting. In this work, we propose a unified bi-level meta-learning framework that enables the EEG decoder to simultaneously perform fast adaptation on future subjects and retain knowledge of previous subjects. In addition, we extend to the more general subject-agnostic scenario and propose a subject shift detection algorithm for situations that subject identity and the occurrence of subject shifts are unknown. We conducted experiments on three public EEG datasets for both subject-aware and subject-agnostic scenarios. The proposed method demonstrates its effectiveness in most of the ablation settings, e.g. an improvement of 5.73% for forgetting mitigation and 3.50% for forward adaptation on SEED dataset for subject agnostic scenarios.
{"title":"Retain and Adapt: Online Sequential EEG Classification With Subject Shift","authors":"Tiehang Duan;Zhenyi Wang;Li Shen;Gianfranco Doretto;Donald A. Adjeroh;Fang Li;Cui Tao","doi":"10.1109/TAI.2024.3385390","DOIUrl":"https://doi.org/10.1109/TAI.2024.3385390","url":null,"abstract":"Large variance exists in Electroencephalogram (EEG) signals with its pattern differing significantly across subjects. It is a challenging problem to perform online sequential decoding of EEG signals across different subjects, where a sequence of subjects arrive in temporal order and no signal data is jointly available beforehand. The challenges include the following two aspects: 1) the knowledge learned from previous subjects does not readily fit to future subjects, and fast adaptation is needed in the process; and 2) the EEG classifier could drastically erase information of learnt subjects as learning progresses, namely catastrophic forgetting. Most existing EEG decoding explorations use sizable data for pretraining purposes, and to the best of our knowledge we are the first to tackle this challenging online sequential decoding setting. In this work, we propose a unified bi-level meta-learning framework that enables the EEG decoder to simultaneously perform fast adaptation on future subjects and retain knowledge of previous subjects. In addition, we extend to the more general subject-agnostic scenario and propose a subject shift detection algorithm for situations that subject identity and the occurrence of subject shifts are unknown. We conducted experiments on three public EEG datasets for both subject-aware and subject-agnostic scenarios. The proposed method demonstrates its effectiveness in most of the ablation settings, e.g. an improvement of 5.73% for forgetting mitigation and 3.50% for forward adaptation on SEED dataset for subject agnostic scenarios.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-05DOI: 10.1109/TAI.2024.3385387
Tingkai Chen;Ning Wang
In this article, to holistically tackle underwater detection degradation due to unknown geometric variation arising from scale, pose, viewpoint, and occlusion under low-contrast and color-distortion circumstances, a shuffled grouping cross-channel attention-based bilateral-filter-interpolation deformable ConvNet (SGCA-BDC) framework is established for benthonic organism detection (BOD). Main contributions are as follows: 1) By comprehensively considering spatial and feature similarities between offset and integral coordinate positions, the BDC with modulation weight mechanism is created, such that sampling ability of convolutional kernel for BO with unknown geometric variation can be adaptively augmented from spatial perspective; 2) By utilizing 1-D convolution to recalibrate channel weight for grouped subfeature via information entropy statistic technique, an SGCA module is innovated, such that seabed background noise can be suppressed from channel aspect; 3) The proposed SGCA-BDC scheme is eventually built in an organic manner by incorporating BDC and SGCA modules. Comprehensive experiments and comparisons demonstrate that the SGCA-BDC scheme remarkably outperforms typical detection approaches including Faster RCNN, SSD, YOLOv6, YOLOv7, YOLOv8, RetinaNet, and CenterNet in terms of mean average precision by 8.54%, 4.4%, 5.18%, 3.1%, 3.01%, 12.53%, and 7.09%, respectively.
{"title":"Shuffled Grouping Cross-Channel Attention-Based Bilateral-Filter-Interpolation Deformable ConvNet With Applications to Benthonic Organism Detection","authors":"Tingkai Chen;Ning Wang","doi":"10.1109/TAI.2024.3385387","DOIUrl":"https://doi.org/10.1109/TAI.2024.3385387","url":null,"abstract":"In this article, to holistically tackle underwater detection degradation due to unknown geometric variation arising from scale, pose, viewpoint, and occlusion under low-contrast and color-distortion circumstances, a shuffled grouping cross-channel attention-based bilateral-filter-interpolation deformable ConvNet (SGCA-BDC) framework is established for benthonic organism detection (BOD). Main contributions are as follows: 1) By comprehensively considering spatial and feature similarities between offset and integral coordinate positions, the BDC with modulation weight mechanism is created, such that sampling ability of convolutional kernel for BO with unknown geometric variation can be adaptively augmented from spatial perspective; 2) By utilizing 1-D convolution to recalibrate channel weight for grouped subfeature via information entropy statistic technique, an SGCA module is innovated, such that seabed background noise can be suppressed from channel aspect; 3) The proposed SGCA-BDC scheme is eventually built in an organic manner by incorporating BDC and SGCA modules. Comprehensive experiments and comparisons demonstrate that the SGCA-BDC scheme remarkably outperforms typical detection approaches including Faster RCNN, SSD, YOLOv6, YOLOv7, YOLOv8, RetinaNet, and CenterNet in terms of mean average precision by 8.54%, 4.4%, 5.18%, 3.1%, 3.01%, 12.53%, and 7.09%, respectively.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-04-05DOI: 10.1109/TAI.2024.3385386
Zongwen Fan;Jin Gou;Shaoyuan Weng
Concrete is a vital component in modern construction, prized for its strength, durability, and versatility. Accurately determining the quantities of concrete components is crucial in civil engineering applications to optimize resources (e.g., manpower and financial resources). In this article, we propose an unbiased fuzzy-weighted relative error support vector machine (UFW-RE-SVM) for reverse prediction of concrete components. First, we add an unbiased term to the target function of UFW-RE-SVM for obtaining an unbiased model. Second, we design a fuzzy-weighted operation to indicate sample importance by incorporating the fuzzy membership values into the UFW-RE-SVM. The $n$ th root operation is introduced to address the exponential explosion issue in the fuzzy-weighted operation. Finally, considering the UFW-RE-SVM is sensitive to its hyperparameters for multioutput prediction, the whale optimization algorithm (WOA) is utilized for hyperparameter optimization for its effectiveness in optimization tasks. We design the fitness function based on the results from multiple components to balance the performance of multioutput predictions. Experimental results show that the performance of our proposed model outperforms existing works in predicting concrete components in terms of mean absolute relative error, standard deviation, and root mean square error. Further, the statistical test shows the WOA and two other metaheuristics can significantly improve the prediction performance. This indicates that the unbiased term, fuzzy-weighted operation, and WOA are effective for improving the proposed model for reverse prediction concrete components. With these promising results, the proposed model could provide decision-makers with a valuable tool for determining concrete component quantities based on desired concrete qualities.
{"title":"An Unbiased Fuzzy Weighted Relative Error Support Vector Machine for Reverse Prediction of Concrete Components","authors":"Zongwen Fan;Jin Gou;Shaoyuan Weng","doi":"10.1109/TAI.2024.3385386","DOIUrl":"https://doi.org/10.1109/TAI.2024.3385386","url":null,"abstract":"Concrete is a vital component in modern construction, prized for its strength, durability, and versatility. Accurately determining the quantities of concrete components is crucial in civil engineering applications to optimize resources (e.g., manpower and financial resources). In this article, we propose an unbiased fuzzy-weighted relative error support vector machine (UFW-RE-SVM) for reverse prediction of concrete components. First, we add an unbiased term to the target function of UFW-RE-SVM for obtaining an unbiased model. Second, we design a fuzzy-weighted operation to indicate sample importance by incorporating the fuzzy membership values into the UFW-RE-SVM. The \u0000<inline-formula><tex-math>$n$</tex-math></inline-formula>\u0000th root operation is introduced to address the exponential explosion issue in the fuzzy-weighted operation. Finally, considering the UFW-RE-SVM is sensitive to its hyperparameters for multioutput prediction, the whale optimization algorithm (WOA) is utilized for hyperparameter optimization for its effectiveness in optimization tasks. We design the fitness function based on the results from multiple components to balance the performance of multioutput predictions. Experimental results show that the performance of our proposed model outperforms existing works in predicting concrete components in terms of mean absolute relative error, standard deviation, and root mean square error. Further, the statistical test shows the WOA and two other metaheuristics can significantly improve the prediction performance. This indicates that the unbiased term, fuzzy-weighted operation, and WOA are effective for improving the proposed model for reverse prediction concrete components. With these promising results, the proposed model could provide decision-makers with a valuable tool for determining concrete component quantities based on desired concrete qualities.","PeriodicalId":73305,"journal":{"name":"IEEE transactions on artificial intelligence","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-04-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142165088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：