检索结果-内蒙古大学图书馆

NPJ HERITAGE SCIENCE 2025年第1期13卷

作者： Zhang, Yu Zhao, Haiying Qi, Lin Zhang, Jian Zhang, Tian Beijing Univ Posts & Telecommun State Key Lab Informat Photon & Opt Commun Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun Sch Artificial Intelligence Beijing 100876 Peoples R China Beijing Informat Sci & Technol Univ Sch Management Sci & Engn Beijing 102206 Peoples R China Beijing Key Lab Green Dev Decis Based Big Data Beijing 100192 Peoples R China Res Ctr Knowledge Management Beijing 100192 Peoples R China

Embroidery is an important cultural heritage that traces the origins of Chinese civilization over five thousand years. Embroidery patterns function as cultural genes characterized by relative stability, inheritance, and variability. Excavating their intrinsic co-occurrence relationships and features can reveal underlying cultural transmission and evolution. By constructing a temporal multilayer network of embroidery pattern co-occurrence, we analyze intra-layer and inter-layer topological properties, identify key pattern nodes, and employ a link clustering algorithm for community detection. The key findings are as follows: (1) The pattern co-occurrence network exhibits scale-free and small-world properties. (2) The pattern co-occurrence network possesses a community structure with divisible and re-combinable functions. (3) The 'Fu-Shou' (fortune-longevity) patterns are the most representative cultural memory symbols of the Qing Dynasty. (4) The fluctuations in network topological properties and pattern node strength centrality are inherently linked to the transitions in social culture and political power.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Recent advances on loss functions in deep learning for computer vision

引用

NEUROCOMPUTING 2022年 497卷 129-158页

作者： Tian, Yingjie Su, Duo Lauria, Stanislao Liu, Xiaohui Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing 100049 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Brunel Univ London Dept Comp Sci London UB8 3PH England Univ Chinese Acad Sci Sch Econ & Management 80 Zhongguancun East Rd Beijing 100190 Peoples R China

The loss function, also known as cost function, is used for training a neural network or other machine learning models. Over the past decade, researchers have designed many loss functions for machine learning, such as mean squared error and mean absolute error. However, in deep learning, neurons of the last layer are usually activated by a sigmoid or softmax function. Thus, training with traditional losses would cause lower efficiency and accuracy. Recently, designing loss functions for deep learning methods has become one of the most challenging problems. This paper provides a comprehensive review of the recent progress and frontiers about loss functions in deep learning, especially for computer vision tasks. Specifically, we discuss the loss functions in three main computer vision tasks, i.e., object detection, face recognition, and image segmentation. Scholars have proposed several novel loss functions to cope with the specific problems such as imbalanced data, uncertain distribution of the predicted bounding boxes, varied overlapped mode between two bounding boxes and over-fitting. The survey details the source, derivation, and properties of each loss function. Furthermore, we also provide some advanced challenges about robust losses, generative adversarial networks, noise-tolerant losses, and semantic data augmentation. Finally, we deliver a summary and some promising future research directions. (c) 2022 Elsevier B.V. All rights reserved.

关键词： Loss function Deep learning Object detection Face recognition Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Unsupervised anomaly segmentation via deep feature reconstruction

引用

NEUROCOMPUTING 2021年 424卷 9-22页

作者： Shi, Yong Yang, Jie Qi, Zhiquan Univ Chinese Acad Sci Sch Econ & Management Beijing 101408 Peoples R China Univ Chinese Acad Sci Sch Comp & Technol Beijing 101408 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Univ Nebraska Coll Informat Sci & Technol Omaha NE 68182 USA

Automatic detecting anomalous regions in images of objects or textures without priors of the anomalies is challenging, especially when the anomalies appear in very small areas of the images, making difficult to-detect visual variations, such as defects on manufacturing products. This paper proposes an effective unsupervised anomaly segmentation approach that can detect and segment out the anomalies in small and confined regions of images. Concretely, we develop a multi-scale regional feature generator which can generate multiple spatial context-aware representations from pre-trained deep convolutional networks for every subregion of an image. The regional representations not only describe the local characteristics of corresponding regions but also encode their multiple spatial context information, making them discriminative and very beneficial for anomaly detection. Leveraging these descriptive regional features, we then design a deep yet efficient convolutional autoencoder and detect anomalous regions within images via fast feature reconstruction. Our method is simple yet effective and efficient. It advances the state-of-the-art performances on several benchmark datasets and shows great potential for real applications. (c) 2020 Published by Elsevier B.V.

关键词： Anomaly detection Anomaly segmentation Regional representation Feature reconstruction

来源：评论

学校读者我要写书评

暂无评论

Supplier's goal setting considering sustainability: An uncertain dynamic data Envelopment Analysis based benchmarking model

引用

INFORMATION SCIENCES 2021年 545卷 44-64页

作者： Zhou, Xiaoyang Li, Linzi Wen, Haoyu Tian, Xin Wang, Shouyang Lev, Benjamin Xidian Univ Sch Econ & Management Xian 710126 Peoples R China Chinese Acad Sci Acad Math & Syst Sci Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Drexel Univ LeBow Coll Business Philadelphia PA 19104 USA

Because of the growing awareness of balancing the economy, the society and the environment, there has been an increased focus on goal setting under sustainable context. However, underperforming suppliers may set impractical goals that are difficult to achieve in the short term due to the poor information analysis, which could cause benchmarking difficulties. Therefore, rather than focusing on a single ultimate goal, this paper developed an implementable goal sequence for which a dynamic data Envelopment Analysis based benchmarking model was developed as a multi-period process. The indicators that could not be directly measured using numerical values were incorporated into the model as fuzzy numbers, and based on efficiencies, the suppliers were divided into several layers. A method that identified supplier attractiveness and progress was then introduced to select the most appropriate goal in each layer for the goal sequence. A numerical example is provided that illustrated the feasibility and efficacy of the proposed method to successfully identify the inefficient suppliers and determine a goal sequence that had three attainable goals. The sensitivity analysis and comparisons revealed that the proposed model avoided selecting period-inefficient suppliers and provided a better benchmarking process. (C) 2020 Elsevier Inc. All rights reserved.

关键词： Goal setting Sustainability Uncertain dynamic DEA Fuzzy data

来源：评论

学校读者我要写书评

暂无评论

Optimizing Two-Way Partial AUC With an End-to-End Framework

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2023年第8期45卷 10228-10246页

作者： Yang, Zhiyong Xu, Qianqian Bao, Shilong He, Yuan Cao, Xiaochun Huang, Qingming Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing 100049 Peoples R China Chinese Acad Sci Inst Comp Technol Key Lab Intelligent Informat Proc Beijing 100190 Peoples R China Chinese Acad Sci Inst Informat Engn State Key Lab Informat Secur SKLOIS Beijing 100093 Peoples R China Univ Chinese Acad Sci Sch Cyber Secur Beijing 100049 Peoples R China Alibaba Grp Secur Dept Hangzhou 311121 Zhejiang Peoples R China Sun Yat Sen Univ Sch Cyber Sci & Technol Shenzhen 518107 Guangdong Peoples R China Univ Chinese Acad Sci Sch Comp Sci & Technol Beijing 101408 Peoples R China Univ Chinese Acad Sci Key Lab Big Data Min & Knowledge Management BDKM Beijing 101408 Peoples R China Peng Cheng Lab Shenzhen 518055 Guangdong Peoples R China

The Area Under the ROC Curve (AUC) is a crucial metric for machine learning, which evaluates the average performance over all possible True Positive Rates (TPRs) and False Positive Rates (FPRs). Based on the knowledge that a skillful classifier should simultaneously embrace a high TPR and a low FPR, we turn to study a more general variant called Two-way Partial AUC (TPAUC), where only the region with TPR >= alpha, FPR <= beta is included in the area. Moreover, a recent work shows that the TPAUC is essentially inconsistent with the existing Partial AUC metrics where only the FPR range is restricted, opening a new problem to seek solutions to leverage high TPAUC. Motivated by this, we present the first trial in this article to optimize this new metric. The critical challenge along this course lies in the difficulty of performing gradient-based optimization with end-to-end stochastic training, evenwith a proper choice of surrogate loss. To address this issue, we propose a generic framework to construct surrogate optimization problems, which supports efficient end-to-end training with deep learning. Moreover, our theoretical analyses show that: 1) the objective function of the surrogate problems will achieve an upper bound of the original problem under mild conditions, and 2) optimizing the surrogate problems leads to good generalization performance in terms of TPAUC with a high probability. Finally, empirical studies over several benchmark datasets speak to the efficacy of our framework.

关键词： AUC Optimization machine learning partial AUC

来源：评论

学校读者我要写书评

暂无评论

A Construction of Robust Representations for Small data Sets Using Broad Learning System

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2021年第10期51卷 6074-6084页

作者： Tang, Huimin Dong, Peiwu Shi, Yong Beijing Inst Technol Sch Management & Econ Beijing 100081 Peoples R China Univ Nebraska Coll Informat Sci & Technol Omaha NE 68182 USA Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Southwest Minzu Univ Coll Elect & Informat Engn Chengdu 610041 Peoples R China

Feature processing is an important step for modeling and can improve the accuracy of machine learning models. Feature extraction methods can effectively extract features from high-dimensional data sets and enhance the accuracy of tasks. However, the performance of feature extraction methods is not stable in low-dimensional data sets. This article extends the broad learning system (BLS) to a framework for constructing robust representations in low-dimensional and small data sets. First, the BLS changed from a supervised prediction method to an ensemble feature extraction method. Second, feature extraction methods instead of random mapping are used to generate mapped features. Third, deep representations, called enhancement features, are learned from the ensemble mapped features. Fourth, data for generating mapped features and enhancement features can be randomly selected. The ensemble of mapped features and enhancement features can provide robust representations to enhance the performance of downstream tasks. A label-based autoencoder (LA) is embedded in the BLS framework as an example to show the effectiveness of the framework. A random LA (RLA) is presented to generate more different features. The experimental results show that the BLS framework can construct robust representations and significantly promote the performance of machine learning models.

关键词： Feature extraction Machine learning Zinc Neural networks Learning systems Task analysis data models Broad learning system (BLS) feature extraction label-based autoencoder (LA) random LA (RLA) robust representation

来源：评论

学校读者我要写书评

暂无评论

Sparse and semi-attention guided faults diagnosis approach for distributed online services

引用

APPLIED SOFT COMPUTING 2023年 148卷

作者： Zhang, Linzi Shi, Yong Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Univ Birmingham Sch Comp Sci Birmingham B15 2TT England Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100049 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China

Despite the rapid advance of unsupervised reconstruction models in online service fault diagnosis, existing methods still lead to frequent false positive or false negative alarms. Tracing to the cause, popular reconstruction frameworks may fall into an "identical shortcut'', where both normal and anomalous cases can be well-recovered and hence faults are escaped from monitoring. Moreover, regularly "concept drift"deployed by online services always leads to sequences distribution dramatically change, which is easily wrongly-detected as anomalies. In this work, we present SS-Attention to avoid two contradictory false alarms simultaneously in a unified framework. Specifically, we first revisit the frameworks of MLP, CNN, and transformer, and confirm the important role of self-attention in preventing the model from learning shortcut, which we therefore select as our backbone. Second, we come up with a sparse attention layer to help model capture more long-term dependencies and distinguish local "concept drift"from anomalies. Third, we propose a semi-attention layer that urges model only learn from normal regions and further avoid well-recovering the faults. We evaluate our model on three public online service datasets and surpass the state-of-the-art alternatives by a sufficiently large margin, achieving an average 95.91% F1-Score with an approx. 34.48% economy of training time.

关键词： Online service systems Multivariate time series key Performance Indicators (KPIs) Semi-attention layers Sparse attention

来源：评论

学校读者我要写书评

暂无评论

The neural network methods for solving Traveling Salesman Problem 8

The neural network methods for solving Traveling Salesman Pr...

引用

8th International Conference on Information Technology and Quantitative management (ITQM) - Developing Global Digital Economy after COVID-19

作者： Shi, Yong Zhang, Yuanying Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Univ Nebraska Omaha Coll Informat Sci & Technol Omaha NE 68182 USA Univ Chinese Acad Sci Sch Math Sci Beijing 100190 Peoples R China

Traveling Salesman Problem(TSP) is a main attention issue at present. Neural network can be used to solve combinatorial optimization problems. In recent years, there have existed many neural network methods for solving TSP, which has made a big step forward for solving combinatorial optimization problems. This paper reviews the neural network methods for solving TSP in recent years, including Hopfield neural network, graph neural network and neural network with reinforcement learning. Using neural network to solve TSP can effectively improve the accuracy of the approximate solution. Finally, we put forward the prospect of solving TSP in the future. (C) 2021 The Authors. Published by Elsevier B.V.

关键词： Traveling Salesman Problem Neural network Hopfield neural network Graph neural network Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Sparse optimization guided pruning for neural networks

引用

NEUROCOMPUTING 2024年 574卷

作者： Shi, Yong Tang, Anda Niu, Lingfeng Zhou, Ruizhi Univ Chinese Acad Sci Sch Math Sci Beijing 100190 Peoples R China Univ Chinese Acad Sci Sch Econ & Management Beijing 100190 Peoples R China Chinese Acad Sci Res Ctr Fictitious Econ & Data Sci Beijing 100190 Peoples R China Chinese Acad Sci Key Lab Big Data Min & Knowledge Management Beijing 100190 Peoples R China Beijing Univ Technol Inst Operat Res & Informat Engn Beijing 100124 Peoples R China

Neural network pruning is a critical field aimed at reducing the infrastructure costs of neural networks by removing parameters. Traditional methods follow a fixed paradigm including pretraining, pruning, and fine-tuning. Despite the close relationship among these three stages, most pruning methods treat them as independent processes. In this paper, we propose a novel two-stage pruning method, which includes pretraining a network that is instructive for subsequent pruning, and a unified optimization model that integrates pruning and fine-tuning. Specifically, in the first stage, we design a group sparse regularized model for pretraining. This model not only safeguards the network from irreversible damage but also offers valuable insights for the pruning process. In the second stage, we introduce an element-wise sparse regularization into pruning model. This model enables us to pinpoint sparse weights more precisely than pretrained network. It automatically derives effective pruning criteria, and omits the step of fine-tuning. To implement the two-stage process in practice, we utilize stochastic gradient algorithm for the pretraining and design a threshold algorithm for pruning stage. Extensive experiments confirm the competitive performance of our proposed method in terms of both accuracy and memory cost when compared to various benchmarks. Furthermore, ablation experiments validate the effectiveness of the proposed pretraining model's guidance for the pruning process.

关键词： Deep neural networks Model compression Sparse optimization

来源：评论

学校读者我要写书评

暂无评论

The Development of Green Olympic Games: A Comparison of CO2Emission Reduction Strategies between PyeongChang 2018 and Beijing 2022 9th

The Development of Green Olympic Games: A Comparison of CO2E...

引用

9th International Conference on Information Technology and Quantitative management, ITQM 2022

作者： Ding, Xuesong Jiang, Yilin Yang, Banban Su, Jingchun Chen, Liming Li, Na School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China Chinese Academy of Fiscal Sciences Ministry of Finance Beijing100142 China Division of Sports Science and Physical Education Tsinghua University Beijing100084 China Division of Sports Science and Physical Education Tangshan Polytechnic College Tangshan063299 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy Sciences Beijing100190 China

The Olympic Games are striking a balance between economic and ecological benefits. Taking the 5 ice event venues of the 2018 PyeongChang Winter Olympic Games as the research object, the CO2 emissions under the 2018 PyeongChang Low-carbon scenario and 2022 Beijing Low-carbon scenario are compared. The empirical results show that, in terms of venue operation, the Beijing strategies can reduce CO2 emissions by 715.7 t compared with the PyeongChang strategies. In terms of the refrigerant escape, Beijing's adoption of CO2 refrigerant can reduce the emission of CO2 by 1202.2 t compared with the R404 refrigerant used in PyeongChang. This paper confirms the progress of the Green Winter Olympic Games, and may provide implications for the future host countries as well as the cities which in the green transformation. © 2022 The Author(s).

关键词： Refrigerants

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：