检索结果-内蒙古大学图书馆

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Shilong Bao Qianqian Xu Zhiyong Yang Yuan He Xiaochun Cao Qingming Huang State Key Laboratory of Information Security Institute of Information Engineering CAS and School of Cyber Security University of Chinese Academy of Sciences Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS School of Computer Science and Tech. University of Chinese Academy of Sciences Alibaba Group School of Cyber Science and Technology Shenzhen Campus Sun Yat-sen University Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS and School of Computer Science and Tech. University of Chinese Academy of Sciences and Key Laboratory of Big Data Mining and Knowledge Management CAS and Peng Cheng Laboratory

ISBN: (纸本)9781713871088

Collaborative Metric Learning (CML) has recently emerged as a popular method in recommendation systems (RS), closing the gap between metric learning and Collaborative Filtering. Following the convention of RS, existing methods exploit unique user representation in their model design. This paper focuses on a challenging scenario where a user has multiple categories of interests. Under this setting, we argue that the unique user representation might induce preference bias, especially when the item category distribution is imbalanced. To address this issue, we propose a novel method called Diversity-Promoting Collaborative Metric Learning (DPCML), with the hope of considering the commonly ignored minority interest of the user. The key idea behind DPCML is to include a multiple set of representations for each user in the system. Based on this embedding paradigm, user preference toward an item is aggregated from different embeddings by taking the minimum item-user distance among the user embedding set. Furthermore, we observe that the diversity of the embeddings for the same user also plays an essential role in the model. To this end, we propose a Diversity Control Regularization Scheme (DCRS) to accommodate the multi-vector representation strategy better. Theoretically, we show that DPCML could generalize well to unseen test data by tackling the challenge of the annoying operation that comes from the minimum value. Experiments over a range of benchmark datasets speak to the efficacy of DPCML.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalised Latent Assimilation in Heterogeneous Reduced Spaces with Machine Learning Surrogate Models

arXiv

引用

arXiv 2022年

作者： Cheng, Sibo Chen, Jianhua Anastasiou, Charitos Angeli, Panagiota Matar, Omar K. Guo, Yi-Ke Pain, Christopher C. Arcucci, Rossella Data Science Instituite Department of Computing Imperial College London United Kingdom Department of Chemical Engineering Imperial College London United Kingdom Department of Chemical Engineering University College London United Kingdom Department of Earth Science & Engineering Imperial College London United Kingdom State Key Laboratory of Multiphase Complex Systems Institute of Process Engineering Chinese Academy of Sciences China

Reduced-order modelling and low-dimensional surrogate models generated using machine learning algorithms have been widely applied in high-dimensional dynamical systems to improve the algorithmic efficiency. In this paper, we develop a system which combines reduced-order surrogate models with a novel data assimilation (DA) technique used to incorporate real-time observations from different physical spaces. We make use of local smooth surrogate functions which link the space of encoded system variables and the one of current observations to perform variational DA with a low computational cost. The new system, named Generalised Latent Assimilation can benefit both the efficiency provided by the reduced-order modelling and the accuracy of data assimilation. A theoretical analysis of the difference between surrogate and original assimilation cost function is also provided in this paper where an upper bound, depending on the size of the local training set, is given. The new approach is tested on a high-dimensional computational fluid dynamics (CFD) application of a two-phase liquid flow with non-linear observation operators that current Latent Assimilation methods can not handle. Numerical results demonstrate that the proposed assimilation approach can significantly improve the reconstruction and prediction accuracy of the deep learning surrogate model which is nearly 1000 times faster than the CFD simulation. © 2022, CC BY-NC-SA.

关键词： Computational fluid dynamics

来源：评论

学校读者我要写书评

暂无评论

iGniter: Interference-Aware GPU Resource Provisioning for Predictable DNN Inference in the Cloud

arXiv

引用

arXiv 2022年

作者： Xu, Fei Xu, Jianian Chen, Jiabin Chen, Li Shang, Ruitao Zhou, Zhi Liu, Fangming The Shanghai Key Laboratory of Multidimensional Information Processing School of Computer Science and Technology East China Normal University 3663 N. Zhongshan Road Shanghai200062 China The School of Computing and Informatics University of Louisiana at Lafayette 301 East Lewis Street LafayetteLA70504 United States The Guangdong Key Laboratory of Big Data Analysis and Processing School of Computer Science and Engineering Sun Yat-sen University 132 E. Waihuan Road Guangzhou510006 China The National Engineering Research Center for Big Data Technology and System The Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology 1037 Luoyu Road Wuhan430074 China

GPUs are essential to accelerating the latency-sensitive deep neural network (DNN) inference workloads in cloud datacenters. To fully utilize GPU resources, spatial sharing of GPUs among co-located DNN inference workloads becomes increasingly compelling. However, GPU sharing inevitably brings severe performance interference among co-located inference workloads, as motivated by an empirical measurement study of DNN inference on EC2 GPU instances. While existing works on guaranteeing inference performance service level objectives (SLOs) focus on either temporal sharing of GPUs or reactive GPU resource scaling and inference migration techniques, how to proactively mitigate such severe performance interference has received comparatively little attention. In this paper, we propose iGniter, an interference-aware GPU resource provisioning framework for cost-efficiently achieving predictable DNN inference in the cloud. iGniter is comprised of two key components: (1) a lightweight DNN inference performance model, which leverages the system and workload metrics that are practically accessible to capture the performance interference;(2) A cost-efficient GPU resource provisioning strategy that jointly optimizes the GPU resource allocation and adaptive batching based on our inference performance model, with the aim of achieving predictable performance of DNN inference workloads. We implement a prototype of iGniter based on the NVIDIA Triton inference server hosted on EC2 GPU instances. Extensive prototype experiments on four representative DNN models and datasets demonstrate that iGniter can guarantee the performance SLOs of DNN inference workloads with practically acceptable runtime overhead, while saving the monetary cost by up to 25% in comparison to the state-of-the-art GPU resource provisioning strategies. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Sample hardness based gradient loss for long-tailed cervical cell detection

arXiv

引用

arXiv 2022年

作者： Liu, Minmin Li, Xuechen Gao, Xiangbo Chen, Junliang Shen, Linlin Wu, Huisi Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China Shenzhen Institute of Artificial Intelligence of Robotics of Society Shenzhen China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China University of California Irvine United States

Due to the difficulty of cancer samples collection and annotation, cervical cancer datasets usually exhibit a long-tailed data distribution. When training a detector to detect the cancer cells in a WSI (Whole Slice Image) image captured from the TCT (Thinprep Cytology Test) specimen, head categories (e.g. normal cells and inflammatory cells) typically have a much larger number of samples than tail categories (e.g. cancer cells). Most existing state-of-the-art long-tailed learning methods in object detection focus on category distribution statistics to solve the problem in the long-tailed scenario, without considering the "hardness" of each sample. To address this problem, in this work we propose a Grad-Libra Loss that leverages the gradients to dynamically calibrate the degree of hardness of each sample for different categories, and re-balance the gradients of positive and negative samples. Our loss can thus help the detector to put more emphasis on those hard samples in both head and tail categories. Extensive experiments on a long-tailed TCT WSI image dataset show that the mainstream detectors, e.g. RepPoints, FCOS, ATSS, YOLOF, etc. trained using our proposed Gradient-Libra Loss, achieved much higher (7.8%) mAP than that trained using cross-entropy classification loss. Copyright © 2022, The Authors. All rights reserved.

关键词： Cytology

来源：评论

学校读者我要写书评

暂无评论

Efficient Estimation of Sensor Biases for the 3-Dimensional Asynchronous Multi-Sensor System

arXiv

引用

arXiv 2021年

作者： Pu, Wenqiang Liu, Ya-Feng Luo, Zhi-Quan Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen518172 China State Key Laboratory of Scientific and Engineering Computing Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China

An important preliminary procedure in multi-sensor data fusion is sensor registration, and the key step in this procedure is to estimate sensor biases from their noisy measurements. There are generally two difficulties in this bias estimation problem: one is the unknown target states which serve as the nuisance variables in the estimation problem, and the other is the highly nonlinear coordinate transformation between the local and global coordinate systems of the sensors. In this paper, we focus on the 3-dimensional asynchronous multi-sensor scenario and propose a weighted nonlinear least squares (NLS) formulation by assuming that there is a target moving with a nearly constant velocity. We propose two possible choices of the weighting matrix in the NLS formulation, which correspond to classical and weighted NLS estimation and maximum likelihood (ML) estimation, respectively. To address the intrinsic nonlinearity, we propose a block coordinate descent (BCD) algorithm for solving the formulated problem, which alternately updates different kinds of bias estimates. Specifically, the proposed BCD algorithm involves solving linear LS problems and nonconvex quadratically constrained quadratic program (QCQP) problems with special structures. Instead of adopting the semidefinite relaxation technique, we develop a much more computationally efficient algorithm based on the alternating direction method of multipliers (ADMM) to solve the nonconvex QCQP subproblems. The convergence of the ADMM to the global solution of the QCQP subproblems is established under mild conditions. The effectiveness and efficiency of the proposed BCD algorithm are demonstrated via numerical simulations. Copyright © 2021, The Authors. All rights reserved.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

Cloud-assisted Road Condition Monitoring with Privacy Protection in VANETs

Cloud-assisted Road Condition Monitoring with Privacy Protec...

引用

International Conference on Mobile Ad-hoc and Sensor Networks, MSN

作者： Lemei Da Yujue Wang Yong Ding Bo Qin Xiaochun Zhou Hai Liang Huiyong Wang Guangxi Key Laboratory of Cryptography and Information Security School of Computer Science and Information Security Guilin University of Electronic Technology Guilin China Pengcheng Laboratory Cyberspace Security Research Center Shenzhen China Key Laboratory of Data Engineering and Knowledge Engineering Ministry of Education School of Information Renmin University of China Beijing China Shenzhen Olym Information Security Technology Co. Ltd. Shenzhen China School of Mathematics and Computing Science Guilin University of Electronic Technology Guilin China

Vehicular ad hoc network (VANET) is one of the fastest developing technologies in intelligent transportation systems (ITS), which has made great contributions to improving traffic congestion and reducing traffic accidents. As it is deployed in an open environment, security and privacy are threatened to a certain extent. Moreover, there are huge data exchanges in high traffic areas, which require VANET system to improve computing efficiency while ensuring communication security. To solve the above issues, this paper proposes a cloud-assisted road condition monitoring (RCM) system. The trusted authority (TA) monitors the road conditions with the help of the cloud server. The vehicle collects the road condition information of the road section managed by the roadside unit (RU), and only the vehicles authorized by the administrative roadside unit can successfully upload the road condition reports to the cloud server. The cloud server divides the road condition reports into different equivalence classes, in this way to report the emergency to the TA when the reported quantity exceeds the threshold. Security analysis showed that the proposed RCM system can effectively protect the security and privacy of road condition reports in VANETs.

关键词： Condition monitoring Privacy data privacy Roads Vehicular ad hoc networks Forgery Sensors

来源：评论

学校读者我要写书评

暂无评论

Mdpnet:A Dual-Pathway Parallel Fusion Network for Multi-Modal MRI Glioma Genotyping

SSRN

引用

SSRN 2024年

作者： Wang, Huaizhi Liu, Haichao Du, Fang Wang, Di Song, Lijuan School of Information Engineering Ningxia University Ningxia Hui Autonomous Region Yinchuan750021 China School of Advanced Interdisciplinary Studies Ningxia University Ningxia Hui Autonomous Region Zhongwei755000 China Ningxia Key Laboratory of Artificial Intelligence and Information Security for Channeling Computing Resources from the East to the West Ningxia Yinchuan750021 China Collaborative Innovation Center for Ningxia Big Data and Artificial Intelligence Co-founded by Ningxia Municipality and Ministry Ningxia Yinchuan750021 China

Glioblastomas have become the most lethal brain tumors in humans, and accurate diagnosis of gliomas is crucial for patient treatment and prognosis. Magnetic Resonance Imaging (MRI) is widely used due to its non-invasiveness and convenience. However, due to the significant heterogeneity of gliomas, even experienced neuroradiologists find it challenging to accurately differentiate glioma types based on imaging phenotypes. According to the WHO classification of central nervous system tumors, the types of gliomas can be determined based on the status of IDH, 1p/19q, and ATRX genetic markers. Current research is mostly limited to the analysis of a single molecular marker, either IDH or 1p/19q, lacking a comprehensive model to simultaneously evaluate these three critical molecular markers. Moreover, the lesion information embedded in T2 and FLAIR modalities has not been fully utilized, thus limiting the accuracy of glioma *** address this issue, we designed a dual-pathway parallel fusion network for multi-modal MRI glioma genotyping (MDPNet), which can simultaneously genotype IDH, 1p/19q, and ATRX genetic markers. MDPNet mainly consists of two parts: attention-guided asymmetric dual-pathway feature extraction (ADFE) and inter-gene relationship-guided genotyping (IGT). ADFE integrates CNN and attention mechanisms to capture both global and local information, enhancing the information from T2 and FLAIR modalities and extracting complex feature information through two parallel pathways. IGT constructs a complex classifier based on the interrelations among genes to simultaneously predict IDH mutation status, 1p/19q co-deletion status, and ATRX mutation *** evaluated our model on public datasets, and experimental results showed that our proposed method exhibited excellent performance. Ablation experiments demonstrated that utilizing the typing information of IDH and ATRX not only improved the network's performance in typing the 1p/19q gene molecule but also furt

关键词： Genes

来源：评论

学校读者我要写书评

暂无评论

GradSA: Gradient Sparsification and Accumulation for Communication-Efficient Distributed Deep Learning 15th

GradSA: Gradient Sparsification and Accumulation for Communi...

引用

15th International Conference on Green, Pervasive, and Cloud computing, GPC 2020

作者： Liu, Bo Jiang, Wenbin Zhao, Shaofeng Jin, Hai He, Bingsheng National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Lab Cluster and Grid Computing Lab Huazhong University of Science and Technology Wuhan China Wuhan National Laboratory for Optoelectronics Key Laboratory of Information Storage System Engineering Research Center of Data Storage Systems and Technology Huazhong University of Science and Technology Wuhan China Library Henan University of Economics and Law Zhengzhou China Department of Computer Science School of Computing National University of Singapore Singapore Singapore

ISBN: (纸本)9783030642426

Large-scale distributed deep learning is of great importance in various applications. For distributed training, the inter-node gradient communication often becomes the performance bottleneck. Gradient sparsification has been proposed to reduce the communication overhead. However, the sparsification only arbitrarily selects a small fraction of gradients, while the efficiency of selection dimension has been overlooked. Furthermore, gradient staleness is inevitable after applying the sparsification which will ultimately lead to model divergence. In this paper, we propose a staleness-compensated sparse stochastic gradient descent algorithm, GradSA, to improve the training efficiency. Layer-level gradients are sparsified to reduce the communication overhead, which conform to the characteristics of the network structure, and historical accumulation of the approximated gradients is utilized to speed up convergence. We demonstrate the model convergence acceleration and the efficiency of our layer-level selection over existing state-of-the-art works, such as DGC, TernGrad, and 8-Bit quantization. © 2020, Springer Nature Switzerland AG.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Self-supervised 3D Point Cloud Completion via Multi-view Adversarial Learning

arXiv

引用

arXiv 2024年

作者： Wu, Lintai Cheng, Xianjing Xu, Yong Zeng, Huanqiang Hou, Junhui Bio-Computing Research Center Harbin Institute of Technology Shenzhen Guangdong Shenzhen518055 China The Department of Computer Science City University of Hong Kong Hong Kong School of Computer Science and Technology Harbin Institute of Technology Shenzhen Guangdong Shenzhen518055 China Department of Computer Science City University of Hong Kong Hong Kong Shenzhen Key Laboratory of Visual Object Detection and Recognition Guangdong Shenzhen518055 China School of Engineering Huaqiao University Quanzhou362021 China School of Information Science and Engineering Huaqiao University Xiamen361021 China

In real-world scenarios, scanned point clouds are often incomplete due to occlusion issues. The task of self-supervised point cloud completion involves reconstructing missing regions of these incomplete objects without the supervision of complete ground truth. Current self-supervised methods either rely on multiple views of partial observations for supervision or overlook the intrinsic geometric similarity that can be identified and utilized from the given partial point clouds. In this paper, we propose MAL-SPC, a framework that effectively leverages both region-level and category-specific geometric similarities to complete missing structures. Our MAL-SPC does not require any 3D complete supervision and only necessitates a single partial point cloud for each object. Specifically, we first introduce a Pattern Retrieval Network to retrieve similar position and curvature patterns between the partial input and the predicted shape, then leverage these similarities to densify and refine the reconstructed results. Additionally, we render the reconstructed complete shape into multi-view depth maps and design an adversarial learning module to learn the geometry of the target shape from category-specific single-view depth images. To achieve anisotropic rendering, we design a density-aware radius estimation algorithm to improve the quality of the rendered images. Our MAL-SPC yields the best results compared to current state-of-the-art methods. We will make the source code publicly available at https://***/ltwu6/malspc. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

How to respond to emerging threats of cyberspace security via social engineering? A novel paradigm to solve cyberspace security issues

TechRxiv

引用

TechRxiv 2022年

作者： Lin, Hao Wang, Chun-Dong Gao, Hao-Yu School of Computer Science and Engineering Tianjin University of Technology Tianjin China Tianjin Key Laboratory of Intelligence Computing Novel Software Technology Ministry of Education Tianjin China Engineering Research Center of Learning-Based Intelligent System Ministry of Education Tianjin China College of Data Science and Application Inner Mongolia University of Technology Hohhot China

Cybersecurity researchers always ignore the "people" behind the issues when dealing with cyberspace security issues. The human and technological aspects of cyberspace security must be simultaneously addressed to enable a secure cyber environment. Social engineering is a discipline whose research object is people, and the combination of social engineering and cyberspace security can provide new solution ideas for lots of cyberspace security issues. This paper attempts to answer the question —— "How to solve cyberspace security issues via social engineering?" We briefly review existing frameworks for social engineering attacks and propose a novel, recursive, and iterative framework for social engineering attacks in cyberspace. The existing research achievements of combination of social engineering and cyberspace security were reviewed in the order of cyber personality, hacker psychology, human flesh search, and social engineering defense. To answer this question, we summarized four "guidelines" for using social engineering. Finally, we look ahead, give our thoughts, and discuss answers of the question and future research. It is aimed at getting more researchers of cyberspace security and social science to pay attention to this field. © 2022, CC BY.

关键词： Cybersecurity

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：