检索结果-内蒙古大学图书馆

Twin Contrastive Learning with Noisy labels

学校读者我要写书评

暂无评论

Twin Contrastive Learning with Noisy Labels

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Zhizhong Huang Junping Zhang Hongming Shan Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Institute of Science and Technology for Brain-inspired Intelligence and MOE Frontiers Center for Brain Science Fudan University Shanghai China Shanghai Center for Brain Science and Brain-inspired Technology Shanghai China

Learning from noisy data is a challenging task that sig-nificantly degenerates the model performance. In this paper, we present TCL, a novel twin contrastive learning model to learn robust representations and handle noisy labels for classification. Specifically, we construct a Gaussian mixture model (GMM) over the representations by injecting the supervised model predictions into GMM to link label- free latent variables in GMM with label-noisy annotations. Then, TCL detects the examples with wrong labels as the out- of-distribution examples by another two-component GMM, taking into account the data distribution. We further propose a cross-supervision with an entropy regularization loss that bootstraps the true targets from model predictions to handle the noisy labels. As a result, TCL can learn discriminative representations aligned with estimated labels through mixup and contrastive learning. Extensive experimental results on several standard benchmarks and real-world datasets demonstrate the superior performance of TCL. In particular, TCL achieves 7.5% improvements on CIFAR-10 with 90% noisy label-an extremely noisy scenario. The source code is available at https://***/Hzzone/TCL.

关键词：

Decoding Continuous Character-based Language from Non-invasive Brain Recordings

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Cenyuan Zheng, Xiaoqing Yin, Ruicheng Geng, Shujie Xu, Jianhan Gao, Xuan Lv, Changze Ling, Zixuan Huang, Xuanjing Cao, Miao Feng, Jianfeng School of Computer Science Fudan University Shanghai China Shanghai Key Laboratory of Intelligent Information Processing Fudan University Shanghai China Institute of Science and Technology for Brain-Inspired Intelligence Fudan University Shanghai China Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence Fudan University Ministry of Education China

Deciphering natural language from brain activity through non-invasive devices remains a formidable challenge. Previous non-invasive decoders either require multiple experiments with identical stimuli to pinpoint cortical regions and enhance signal-to-noise ratios in brain activity, or they are limited to discerning basic linguistic elements such as letters and words. We propose a novel approach to decoding continuous language from single-trial non-invasive fMRI recordings, in which a three-dimensional convolutional network augmented with information bottleneck is developed to automatically identify responsive voxels to stimuli, and a character-based decoder is designed for the semantic reconstruction of continuous language characterized by inherent character structures. The resulting decoder can produce intelligible textual sequences that faithfully capture the meaning of perceived speech both within and across subjects, while existing decoders exhibit significantly inferior performance in cross-subject contexts. The ability to decode continuous language from single trials across subjects demonstrates the promising applications of non-invasive language brain-computer interfaces in both healthcare and neuroscience. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

Geometry-Aware Reference Synthesis for Multi-View Image Super-Resolution

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Cheng, Ri Sun, Yuqi Yan, Bo Tan, Weimin Ma, Chenxi School of Computer Science Shanghai Key Laboratory of Intelligent Information Processing Shanghai Collaborative Innovation Center of Intelligent Visual Computing Fudan University Shanghai China

Recent multi-view multimedia applications struggle between high-resolution (HR) visual experience and storage or bandwidth constraints. Therefore, this paper proposes a Multi-View Image Super-Resolution (MVISR) task. It aims to increase the resolution of multiview images captured from the same scene. One solution is to apply image or video super-resolution (SR) methods to reconstruct HR results from the low-resolution (LR) input view. However, these methods cannot handle large-angle transformations between views and leverage information in all multi-view images. To address these problems, we propose the MVSRnet, which uses geometry information to extract sharp details from all LR multi-view to support the SR of the LR input view. Specifically, the proposed Geometry-Aware Reference Synthesis module in MVSRnet uses geometry information and all multi-view LR images to synthesize pixel-aligned HR reference images. Then, the proposed Dynamic High-Frequency Search network fully exploits the high-frequency textural details in reference images for SR. Extensive experiments on several benchmarks show that our method significantly improves over the state-of-the-art approaches. © 2022, CC BY.

关键词： Geometry

SIAM: A Simple Alternating Mixer for Video Prediction

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zheng, Xin Peng, Ziang Cao, Yuan Shan, Hongming Zhang, Junping Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Department of Aeronautics and Astronautics Fudan University Shanghai200433 China Institute of Science and Technology for Brain-inspired Intelligence Fudan University Shanghai200433 China

Video prediction, predicting future frames from the previous ones, has broad applications such as autonomous driving and weather forecasting. Existing state-of-the-art methods typically focus on extracting either spatial, temporal, or spatiotemporal features from videos. Different feature focuses, resulting from different network architectures, may make the resultant models excel at some video prediction tasks but perform poorly on others. Towards a more generic video prediction solution, we explicitly model these features in a unified encoder-decoder framework and propose a simple alternating Mixer (SIAM). The novelty of SIAM lies in the design of dimension alternating mixing (DaMi) blocks, which can model spatial, temporal, and spatiotemporal features through alternating the dimensions of the feature maps. Extensive experimental results demonstrate the superior performance of the proposed SIAM on four benchmark video datasets covering both synthetic and real-world scenarios. Copyright © 2023, The Authors. All rights reserved.

关键词： Benchmarking

Adaptive Nonlinear Latent Transformation for Conditional Face Editing

学校读者我要写书评

暂无评论

Adaptive Nonlinear Latent Transformation for Conditional Fac...

International Conference on computer Vision (ICCV)

作者： Zhizhong Huang Siteng Ma Junping Zhang Hongming Shan Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Institute of Science and Technology for Brain-Inspired Intelligence and MOE Frontiers Center for Brain Science Fudan University Shanghai China Shanghai Center for Brain Science and Brain-Inspired Technology Shanghai China

Recent works for face editing usually manipulate the latent space of StyleGAN via the linear semantic directions. However, they usually suffer from the entanglement of facial attributes, need to tune the optimal editing strength, and are limited to binary attributes with strong supervision signals. This paper proposes a novel adaptive nonlinear latent transformation for disentangled and conditional face editing, termed AdaTrans. Specifically, our AdaTrans divides the manipulation process into several finer steps; i.e., the direction and size at each step are conditioned on both the facial attributes and the latent codes. In this way, AdaTrans describes an adaptive nonlinear transformation trajectory to manipulate the faces into target attributes while keeping other attributes unchanged. Then, AdaTrans leverages a predefined density model to constrain the learned trajectory in the distribution of latent codes by maximizing the likelihood of transformed latent code. Moreover, we also propose a disentangled learning strategy under a mutual information framework to eliminate the entanglement among attributes, which can further relax the need for labeled data. Consequently, AdaTrans enables a controllable face editing with the advantages of disentanglement, flexibility with non-binary attributes, and high fidelity. Extensive experimental results on various facial attributes demonstrate the qualitative and quantitative effectiveness of the proposed AdaTrans over existing state-of-the-art methods, especially in the most challenging scenarios with a large age gap and few labeled examples. The source code is available at https://***/Hzzone/AdaTrans.

关键词：

PGUD: Personalized Graph Universal Defense 29

学校读者我要写书评

暂无评论

PGUD: Personalized Graph Universal Defense

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Gan, Zeming Su, Linlin Li, Xianxian Wang, Jinyan Guangxi Normal University Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guilin China Guangxi Normal University Guangxi Key Lab of Multi-Source Information Mining and Security Guilin China Guangxi Normal University School of Computer Science and Engineering Guilin China

ISBN: (纸本)9798350330717

Extensive evidence shows that Graph Neural Networks (GNNs) are vulnerable to adversarial attacks. Previous work has made great efforts to improve the performance and usability of GNNs, but they remain sensitive and vulnerable to local attacks. Local attacks slightly perturb the target nodes, resulting in misclassification of GNNs on these specific nodes. Graph Universal Defense (GUD) is proposed to generate a defense patch to protect important nodes in a graph from local attacks. However, the research on GUD is still in the preliminary stage. The unique patch does not adaptively protect different nodes, and it has weak robustness against strong local attacks. To address this problem, we propose Personalized Graph Universal Defense (PGUD). It maps nodes into several clusters, each corresponding to a personalized defense patch. We further instantiate a class-specific PGUD. From a perspective of effective perturbation at minimal cost, we theoretically derive the potential attacker nodes manipulated by the adversary, and use patches to mitigate their impact on target nodes. Experimental results show that our approach can be efficiently integrated into any GNN for downstream tasks and provide effective protection for different target nodes in a personalized manner. Our method significantly improves the robustness against local attacks compared to prior work and can be easily scaled to large graphs. © 2023 IEEE.

关键词： adversarial attack adversarial defense Graph neural networks robustness

Heterophilic Graph Representation Learning Based on Multi-Order information Extraction and High and Low Pass Filters

学校读者我要写书评

暂无评论

Heterophilic Graph Representation Learning Based on Multi-Or...

Smart World Congress (SWC), IEEE

作者： Ling Wu Xinyu Li Yingjie Yang Kun Guo Qishan Zhang College of Computer and Data Science Fuzhou University Fuzhou China Fujian Key Laboratory of Network Computing and Intelligent Information Processing China School of Computer Science and Informatics De Montfort University Leicester England College of Economics and Humanities Shanghai International Studies University Shanghai China

ISBN: (数字)9798331520861

ISBN: (纸本)9798331520878

Real-world networks can be divided into homophilic and heterophilic graphs. Significantly, many nodes tend to be heterophilic in a homophilic graph. Graph representation learning in heterophilic graphs has attracted considerable significance in recent years as it breaks the homophily assumption and achieves higher-quality graph embeddings. However, existing heterophilic graph representation learning methods often mix attribute and topological information, neglecting the mutual exclusion between attribute information and topological information. Furthermore, algorithms in heterogeneous graphs usually rely on specific matrix structures to capture high-order information, which may restrict the applicability of these methods. To tackle these issues, this paper proposes a GNN-based strategy for Heterophilic graph Representation Learning based on Multi-order information extraction and High and Low pass filters (HRLMHL). First, we designed two relatively independent modules to learn attribute information and topological information, respectively, to address the issue of mismatch. Second, we built a multi-order information extraction module to extract multi-order similarity features, thereby achieving the learning of high-order graph information. Experiments on real datasets show that the node classification accuracy is superior to that of the state-of-the-art methods on most datasets.

关键词： Representation learning Accuracy Low-pass filters information retrieval Feature extraction Graph neural networks Topology Data mining Unsupervised learning

Measures and Optimization for Robustness and Vulnerability in Disconnected Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhu, Liwang Bao, Qi Zhang, Zhongzhi The Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China The Research Institute of Intelligent Complex Systems Fudan University Shanghai200433 China The Shanghai Engineering Research Institute of Blockchain Shanghai200433 China

The function or performance of a network is strongly dependent on its robustness, quantifying the ability of the network to continue functioning under perturbations. While a wide variety of robustness metrics have been proposed, they have their respective limitations. In this paper, we propose to use the forest index as a measure of network robustness, which overcomes the deficiencies of existing metrics. Using such a measure as an optimization criterion, we propose and study the problem of breaking down a network by attacking some key edges. We show that the objective function of the problem is monotonic but not submodular, which impose more challenging on the problem. We thus resort to greedy algorithms extended for non-submodular functions by iteratively deleting the most promising edges. We first propose a simple greedy algorithm with a proved bound for the approximation ratio and cubic-time complexity. To confront the computation challenge for large networks, we further propose an improved nearly-linear time greedy algorithm, which significantly speeds up the process for edge selection but sacrifices little accuracy. Extensive experimental results for a large set of real-world networks verify the effectiveness and efficiency of our algorithms, demonstrating that our algorithms outperform several baseline schemes. Copyright © 2023, The Authors. All rights reserved.

关键词： Iterative methods

Relaxed Graph Semi-Supervised Contrastive Learning for Node Classification 29

学校读者我要写书评

暂无评论

Relaxed Graph Semi-Supervised Contrastive Learning for Node ...

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Li, Qiyu Li, Xianxian Qian, Haodong Li, De Wang, Jinyan Guangxi Normal University Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guilin China Guangxi Normal University Guangxi Key Lab of Multi-Source Information Mining and Security Guilin China Guangxi Normal University School of Computer Science and Engineering Guilin China

ISBN: (纸本)9798350330717

Graph Neural Networks (GNNs) have emerged as promising tools in graph semi-supervised learning. They acquire low-dimensional node embeddings for downstream tasks by aggregating and updating features from neighboring nodes. However, in a semi-supervised setting, the availability of labeled node information is limited, resulting in the underutilization of vast amounts of unlabeled node information. Current research integrates contrastive learning with graph semi-supervised learning to better use this unlabeled node information. Nevertheless, the differences between cross-entropy and self-supervised contrastive loss are often overlooked. To address this issue, we propose a novel relaxed graph contrastive semi-supervised learning method. We analyze the difference between nodes correctly classified under cross-entropy loss and self-supervised contrastive loss. Inspired by this, we design a relaxed contrastive loss that considers the node features optimized under these two loss functions, thereby expanding the positive sample set. Furthermore, to ensure the quality of the positive sample set, we introduce a threshold constraint based on feature similarity and prediction results to select more reliable positive samples. The experimental results indicate that our method exhibits competitive performance across most datasets. © 2023 IEEE.

关键词： Contrastive learning Graph Neural Networks Semi-supervised learning