检索结果-内蒙古大学图书馆

GT-CAM: Game Theory Based Class Activation Map for GCN

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2024年第12期46卷 8806-8819页

作者： Li, Yanshan Shi, Ting Chen, Zhiyuan Zhang, Li Xie, Weixin Inst Intelligent Informat Proc Guangdong Key Lab Intelligent Informat Proc Shenzhen 518000 Guangdong Peoples R China

Graph Convolutional Networks (GCN) have shown outstanding performance in skeleton-based behavior recognition. However, their opacity hampers further development. Researches on the explainability of deep learning have provided solutions to this issue, with Class Activation Map (CAM) algorithms being a class of explainable methods. However, existing CAM algorithms applies to GCN often independently compute the contribution of individual nodes, overlooking the interactions between nodes in the skeleton. Therefore, we propose a game theory based class activation map for GCN (GT-CAM). First, GT-CAM integrates Shapley values with gradient weights to calculate node importance, producing an activation map that highlights the critical role of nodes in decision-making. It also reveals the cooperative dynamics between nodes or local subgraphs for a more comprehensive explanation. Second, to reduce the computational burden of Shapley values, we propose a method for calculating Shapley values of node coalitions. Lastly, to evaluate the rationality of coalition partitioning, we propose a rationality evaluation method based on bipartite game interaction and cooperative game theory. Additionally, we introduce an efficient calculation method for the coalition rationality coefficient based on the Monte Carlo method. Experimental results demonstrate that GT-CAM outperforms other competitive interpretation methods in visualization and quantitative analysis.

关键词： Action recognition game theory human skeleto Monte Carlo sampling spatial-temporal graph convolution network

来源：评论

学校读者我要写书评

暂无评论

A data-efficient strategy for building high-performing medical foundation models

引用

NATURE BIOMEDICAL ENGINEERING 2025年第4期9卷 1-13页

作者： Sun, Yuqi Tan, Weimin Gu, Zhuoyao He, Ruian Chen, Siyuan Pang, Miao Yan, Bo Fudan Univ Sch Comp Sci Shanghai Key Lab Intelligent Informat Proc Shanghai Peoples R China

Foundation models are pretrained on massive datasets. However, collecting medical datasets is expensive and time-consuming, and raises privacy concerns. Here we show that synthetic data generated via conditioning with disease labels can be leveraged for building high-performing medical foundation models. We pretrained a retinal foundation model, first with approximately one million synthetic retinal images with physiological structures and feature distribution consistent with real counterparts, and then with only 16.7% of the 904,170 real-world colour fundus photography images required in a recently reported retinal foundation model (RETFound). The data-efficient model performed as well or better than RETFound across nine public datasets and four diagnostic tasks;and for diabetic-retinopathy grading, it used only 40% of the expert-annotated training data used by RETFound. We also support the generalizability of the data-efficient strategy by building a classifier for the detection of tuberculosis on chest X-ray images. The text-conditioned generation of synthetic data may enhance the performance and generalization of medical foundation models.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

Poisoning Topology View in Software-Defined Vehicular Network: An Empirical Study

引用

IEEE TRANSACTIONS ON intelligent TRANSPORTATION SYSTEMS 2024年第11期25卷 16805-16816页

作者： Deng, Shuhua Zhang, Xingyu Gao, Xieping Xiangtan Univ Sch Comp Sci Key Lab Intelligent Comp & Informat Proc Minist Educ Xiangtan 411105 Peoples R China Xiangtan Univ Sch Cyberspace Sci Key Lab Intelligent Comp & Informat Proc Minist Educ Xiangtan 411105 Peoples R China Xiangtan Univ Key Lab Intelligent Comp & Informat Proc Minist Educ Xiangtan 411105 Peoples R China Hunan Normal Univ Key Lab Comp & Stochast Math Minist Educ Changsha 410081 Peoples R China

The development of the vehicular ad-hoc network (VANET) provides a promising solution to promoting road safety and driving experiences, but it also generates massive data, leading to network configuration and management issues. Fortunately, by integrating software-defined network (SDN) and VANET, a new network paradigm called software-defined vehicular network (SDVN) is proposed to tackle these problems via furnishing centralized control and programmability. With the help of SDN, the centralized controller provides global visibility about network devices and improves the efficiency of various applications. However, the building procedure of global topology also brings new security concerns to the SDVN. In this paper, we comprehensively investigate the security of topology management under a standard SDVN scenario. By exploiting the high mobility of VANET and vulnerabilities in topology management inherited from SDN, we unveil five threats of topology poison in SDVN with lower attack bars. Based on such threats, we propose several attacks to poison the global topology of four mainstream controllers in emulated and real-world environments. Additionally, we present empirical studies to illustrate the impact of these attacks on network communication and topology-based applications. Finally, we discuss the feasibility of these attacks under existing state-of-the-art defense systems.

关键词： Topology Network topology Vehicular ad hoc networks Toxicology Automobiles Resource management Protocols Software-defined networking VANET topology security

来源：评论

学校读者我要写书评

暂无评论

Subdivision of Adjacent Areas for 3D Point Cloud Semantic Segmentation

引用

SIGNAL IMAGE AND VIDEO procESSING 2025年第1期19卷 1-13页

作者： Xu, Haixia Hu, Kaiyu Xu, Yuting Zhu, Jiang Xiangtan Univ Sch Automat & Elect Informat Key Lab Intelligent Comp & Informat Proc Minist Educ Xiangtan 411100 Peoples R China

In 3D point cloud semantic segmentation, much of the previous research has focused on aggregating the fine-grained geometric structures of local regions, overlooking the long-term features. However, global long-term contextual dependencies play a role as important as local features aggregation. This paper proposes a Subdivision of Adjacent Areas (SAA) module, which efficiently mines more informative features to enrich global long-term contextual dependencies. SAA is constructed by the CoVariance-Enhanced Channel Attention (CECA) and the PseudoNL Spatial Attention (PSA). The former learns the interdependence among channels via second-order statistics for each feature channel, while the latter efficiently captures the positional correlation among points in the entire space via a pseudo feature map. The proposed SAA, a plug-and-play, end-to-end trainable module, can be integrated into existing segmentation networks. Extensive experiments on S3DIS and ScanNet datasets demonstrate that networks integrated with our SAA improve mIoU performance. It verifies that SAA is beneficial for 3D point cloud segmentation networks in achieving excellent performance.

关键词： Semantic segmentation 3D point cloud Global attention Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Transferable Generative Framework for Multi-label Zero-Shot Learning

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第5期34卷 3409-3423页

作者： Ma, Peirong He, Zhiquan Ran, Wu Lu, Hong Fudan Univ Sch Comp Sci Shanghai Key Lab Intelligent Informat Proc Shanghai 200438 Peoples R China

Multi-label zero-shot learning (MLZSL) is a more realistic and challenging task than single-label zero-shot learning (SLZSL), which aims to recognize multiple unseen classes in a single image. To adapt generative models to the MLZSL task and better recognize multiple unseen object categories in an image, this paper proposes a Transferable Generative Framework (TGF), which consists of a Multi-label Semantic Embedding Autoencoders (SEAs), a Semantic-Related Multi-label Feature Transformation Network (FTN) and a Multi-label Feature Generation Networks (FGNs). First, SEAs adaptively encodes the class-level word vectors corresponding to each sample containing different number of classes into sample-level semantic embeddings with the same dimension. Then, FTN transforms global features extracted by a CNN pre-trained on single-label images into features that are semantic-related and more suitable for multi-label classification. Finally, FGNs generates both global and local features to better recognize the dominant and minor object categories in a multi-label image, respectively. Extensive experiments on three benchmark datasets show that TGF significantly outperforms state-of-the-arts. Specifically, compared with the previous best generative MLZSL method (i.e., Gen-MLZSL), TGF improves the mAP of the ZSL (GZSL) task by 5.4% (6.9%), 20.5% (27.9%), and 2.4% (3.9%) on NUS-WIDE, Open Images, and MS-COCO datasets, respectively.

关键词： Multi-label zero-shot learning multi-label semantic embedding autoncoders multi-label feature transformation networks multi-label feature generation networks

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Object-Centric Learning From Multiple Unspecified Viewpoints

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2024年第5期46卷 3897-3909页

作者： Yuan, Jinyang Chen, Tonglin Shen, Zhimeng Li, Bin Xue, Xiangyang Fudan Univ Sch Comp Sci Shanghai Key Lab Intelligent Informat Proc Shanghai 200433 Peoples R China

Visual scenes are extremely diverse, not only because there are infinite possible combinations of objects and backgrounds but also because the observations of the same scene may vary greatly with the change of viewpoints. When observing a multi-object visual scene from multiple viewpoints, humans can perceive the scene compositionally from each viewpoint while achieving the so-called "object constancy" across different viewpoints, even though the exact viewpoints are untold. This ability is essential for humans to identify the same object while moving and to learn from vision efficiently. It is intriguing to design models that have a similar ability. In this article, we consider a novel problem of learning compositional scene representations from multiple unspecified (i.e., unknown and unrelated) viewpoints without using any supervision and propose a deep generative model which separates latent representations into a viewpoint-independent part and a viewpoint-dependent part to solve this problem. During the inference, latent representations are randomly initialized and iteratively updated by integrating the information in different viewpoints with neural networks. Experiments on several specifically designed synthetic datasets have shown that the proposed method can effectively learn from multiple unspecified viewpoints.

关键词： compositional scene representations deep generative models object constancy object-centric learning unsupervised learning variational inference

来源：评论

学校读者我要写书评

暂无评论

Linear Opinion Dynamics Model With Higher Order Interactions

引用

IEEE TRANSACTIONS ON compUTATIONAL SOCIAL SYSTEMS 2024年第3期11卷 3627-3636页

作者： Xu, Wanyue Zhang, Zhongzhi Fudan Univ Sch Comp Sci Shanghai Key Lab Intelligent Informat Proc Shanghai 200433 Peoples R China

Opinion dynamics is a central subject of computational social science, and various models have been developed to understand the evolution and formulation of opinions. Existing models mainly focus on opinion dynamics on graphs that only capture pairwise interactions between agents. In this article, we extend the popular Friedkin-Johnsen model for opinion dynamics on graphs to hypergraphs, which describe higher order interactions occurring frequently on real networks, especially social networks. To achieve this, based on the fact that for linear dynamics, the multiway interactions can be reduced to effective pairwise node interactions, we propose a method to decode the group interactions encoded in hyperedges by undirected edges or directed edges in graphs. We then show that higher order interactions play an important role in the opinion dynamics since the overall steady-state expressed opinion and polarization differ greatly from those without group interactions. We also provide an interpretation of the equilibrium expressed opinion from the perspective of the spanning converging forest, based on which we design a fast sampling algorithm to approximately evaluate the overall opinion and opinion polarization on directed weighted graphs. Finally, we conduct experiments on real-world hypergraph datasets, demonstrating the performance of our algorithm.

关键词： Social networking (online) Mathematical models Collaboration Organizations Laplace equations Forestry Directed graphs computational social science higher order interactions hypergraph opinion dynamics opinion polarization social networks

来源：评论

学校读者我要写书评

暂无评论

Prompt-Guided Semantic-Aware Distillation for Weakly Supervised Incremental Semantic Segmentation

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第11期34卷 10632-10645页

作者： Hao, Xuze Jiang, Xuhao Ni, Wenqian Tan, Weimin Yan, Bo Fudan Univ Shanghai Key Lab Intelligent Informat Proc Sch Comp Sci Shanghai 201203 Peoples R China

Weakly Supervised Incremental Semantic Segmentation (WISS) aims to enable deep neural networks to incrementally learn new classes using only image-level labels without catastrophic forgetting. Despite WISS eliminating the usage of costly and time-consuming pixel-by-pixel annotations, the image-level labels can not provide details about the location of new classes, resulting in inferior performance. To address these issues, we take inspiration from zero-shot learning to model the inter-class semantic relation utilizing class names as text prompts, thereby facilitating knowledge transfer between classes. However, some class names of the segmentation datasets are polysemous. Thus, we design a new prompt template to better capture the semantic relation by appending synonyms and definitions of the corresponding classes. Guided by this semantic relation, we propose semantic relation weighted distillation to transfer the knowledge from old to new classes, significantly improving plasticity while reducing forgetting. Additionally, we introduce a novel superclass-level distillation aimed at preserving shared global knowledge within the superclass, further alleviating catastrophic forgetting. We extensively evaluate our method by integrating it into state-of-the-art WISS approaches on Pascal VOC and COCO datasets. We observe consistent gains in performance across diverse experimental scenarios. Code is available at https://***/Magic-Nova77/PGSD.

关键词： Weakly supervised incremental semantic segmentation knowledge distillation semantic embeddings prompt tuning.

来源：评论

学校读者我要写书评

暂无评论

Gradient Boosting Decision Tree-Based PMM Model Integrated Into FDTD Method for Solving Subsurface Sensing Problems

引用

IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION 2024年第7期72卷 5892-5899页

作者： Feng, Naixing Wang, Huan Zhu, Zixian Zhang, Yuxian Yang, Lixia Huang, Zhixiang Anhui Univ Anhui Higher Educ Inst Anhui Lab Informat Mat & Intelligent Sensing Key Lab Intelligent Comp & Signal ProcKey Lab Ele Hefei 230601 Peoples R China

It is known that both the physical domain in size and problem complexity should also be considered to decrease more and more computational resources, especially for a large-scale complicated problem, besides taking into account numerical accuracy and efficiency for extremely low-frequency subsurface sensing application in the airborne transient electromagnetics (ATEMs). For better satisfying with the above conditions, an alternative perfectly matched monolayer (PMM) model is proposed based on the gradient boosting decision tree (GBDT) for the finite-difference time-domain (FDTD) implementation, which is incorporated to further enhance the performance. In the GBDT-based PMM model, it is not only higher accuracy that could be obtained through the ensemble learning method of feature attention but also less memory and time consumption that are required. In terms of model training stability and its lightweight, moreover, the proposed model has significant merits because of its dependence on the characteristics of traditional machine learning (ML) models. Eventually, numerical cases of low-frequency ATEM applications are considered to validate the performance of the proposed algorithm. It is observed from the results that we can not only gain advantages in accuracy, efficiency, and problem complexity but also succeed in integrating this model into the FDTD solver.

关键词： computational modeling Antennas and propagation Predictive models Sensors Atmospheric modeling Transient analysis Training Airborne transient electromagnetics (ATEMs) finite difference time domain (FDTD) gradient boosting decision tree (GBDT) perfectly matched monolayer (PMM)

来源：评论

学校读者我要写书评

暂无评论

Adaptive Perturbation for Adversarial Attack

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2024年第8期46卷 5663-5676页

作者： Yuan, Zheng Zhang, Jie Jiang, Zhaoyan Li, Liangliang Shan, Shiguang Chinese Acad Sci Inst Comp Technol Key Lab Intelligent Informat Proc Beijing 100190 Peoples R China Chinese Acad Sci Inst Comp Technol Key Lab Intelligent Informat Proc Beijing 100049 Peoples R China Tencent Shenzhen 518057 Peoples R China

In recent years, the security of deep learning models achieves more and more attentions with the rapid development of neural networks, which are vulnerable to adversarial examples. Almost all existing gradient-based attack methods use the sign function in the generation to meet the requirement of perturbation budget on L-infinity norm. However, we find that the sign function may be improper for generating adversarial examples since it modifies the exact gradient direction. Instead of using the sign function, we propose to directly utilize the exact gradient direction with a scaling factor for generating adversarial perturbations, which improves the attack success rates of adversarial examples even with fewer perturbations. At the same time, we also theoretically prove that this method can achieve better black-box transferability. Moreover, considering that the best scaling factor varies across different images, we propose an adaptive scaling factor generator to seek an appropriate scaling factor for each image, which avoids the computational cost for manually searching the scaling factor. Our method can be integrated with almost all existing gradient-based attack methods to further improve their attack success rates. Extensive experiments on the CIFAR10 and ImageNet datasets show that our method exhibits higher transferability and outperforms the state-of-the-art methods.

关键词： Perturbation methods Iterative methods Adaptation models Generators Closed box Security Training Adversarial attack transfer-based attack adversarial example adaptive perturbation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：