检索结果-内蒙古大学图书馆

您好，读者！请登录

内蒙古大学图书馆

首页
概况
党建
资源
服务
科研支持
- 论文收录引用证明
- 科技查新
知识产权
档案馆
帮助

咨询与建议

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

您的常用邮箱：*

您的手机号码：*

问题描述：

当前已输入0个字，您还可以输入200个字

全部搜索
期刊论文
图书
学位论文
标准
纸本馆藏
外文资源发现
数据库导航
超星发现

高级检索

时间限定

出版年份：

文献类型

图书期刊文献学位论文多媒体

馆藏选择

电子馆藏纸本馆藏

核心期刊

全部期刊 SCI 收录期刊 SSCI 收录期刊 EI 收录期刊 CSCD 收录期刊 CSSCI 收录期刊

语言

中文英文

文献类型

期刊文献图书学位论文标准纸本馆藏

帮助

文字说明：

T=题名（书名、题名），A=作者（责任者），K=主题词，P=出版物名称，PU=出版社名称，O=机构（作者单位、学位授予单位、专利申请人），L=中图分类号，C=学科分类号，U=全部字段，Y=年（出版发行年、学位年度、标准发布年）

检索规则说明：

AND代表“并且”；OR代表“或者”；NOT代表“不包含”；(注意必须大写,运算符两边需空一格)

检索范例：

范例一：(K=图书馆学 OR K=情报学) AND A=范并思 AND Y=1982-2016
范例二：P=计算机应用与软件 AND (U=C++ OR U=Basic) NOT K=Visual AND Y=2011-2016

分类表

所选分类

>> <<

限定检索结果

文献类型

1,926 篇 会议
13 篇 期刊文献
10 册 图书

馆藏范围

1,949 篇 电子文献
0 种 纸本馆藏

日期分布

学科分类号

1,033 篇 工学
- 909 篇 计算机科学与技术...
- 491 篇 软件工程
- 297 篇 控制科学与工程
- 122 篇 信息与通信工程
- 116 篇 电气工程
- 73 篇 生物工程
- 44 篇 生物医学工程（可授...
- 40 篇 仪器科学与技术
- 34 篇 机械工程
- 30 篇 电子科学与技术（可...
- 21 篇 化学工程与技术
- 21 篇 网络空间安全
- 13 篇 光学工程
- 9 篇 动力工程及工程热...
321 篇 理学
- 211 篇 数学
- 91 篇 生物学
- 82 篇 统计学（可授理学、...
- 49 篇 系统科学
- 28 篇 物理学
- 17 篇 化学
287 篇 管理学
- 218 篇 管理科学与工程(可...
- 95 篇 图书情报与档案管...
- 32 篇 工商管理
35 篇 医学
- 30 篇 临床医学
- 28 篇 基础医学(可授医学...
- 21 篇 药学(可授医学、理...
21 篇 法学
- 20 篇 社会学
18 篇 经济学
- 18 篇 应用经济学
11 篇 农学
- 11 篇 作物学
7 篇 教育学
- 7 篇 教育学
6 篇 军事学
5 篇 文学
2 篇 艺术学
1 篇 哲学

主题

493 篇 machine learning
165 篇 data mining
155 篇 computer science
153 篇 machine learning...
141 篇 support vector m...
118 篇 cybernetics
105 篇 neural networks
94 篇 clustering algor...
90 篇 computer network...
88 篇 learning systems
80 篇 educational inst...
80 篇 distributed comp...
66 篇 electronic mail
66 篇 artificial intel...
64 篇 support vector m...
62 篇 algorithm design...
60 篇 humans
59 篇 pattern recognit...
58 篇 computational mo...
58 篇 testing

机构

63 篇 department of co...
40 篇 department of co...
24 篇 school of comput...
23 篇 department of co...
14 篇 machine learning...
14 篇 department of co...
14 篇 college of mecha...
13 篇 school of comput...
12 篇 college of compu...
12 篇 institute of com...
11 篇 hebei univ coll ...
8 篇 institute of com...
8 篇 school of comput...
7 篇 department of in...
7 篇 school of comput...
7 篇 department of co...
7 篇 department of au...
7 篇 department of in...
6 篇 key laboratory o...
6 篇 key laboratory o...

作者

26 篇 yeung daniel s.
22 篇 d.s. yeung
16 篇 xi-zhao wang
12 篇 hui wang
12 篇 ng wing w. y.
12 篇 tsang eric c. c.
11 篇 e.c.c. tsang
11 篇 daniel s. yeung
11 篇 kuan-quan wang
10 篇 wang xi-zhao
10 篇 yeung
10 篇 tsang
9 篇 jane you
9 篇 you jane
9 篇 zhang david
8 篇 qing he
8 篇 de-gang chen
7 篇 qing-hua li
7 篇 yu zhiwen
7 篇 li le

语言

1,945 篇 英文
3 篇 中文
1 篇 其他

检索条件"任意字段=International Conference on Machine Learning and Computing（ICMLC 2009）"

共 1949 条记录，以下是21-30 订阅

全选清除本页清除全部题录导出标记到"检索档案"

详细简洁

排序：

Efficient 3D Reconstruction of Multiple Plants from UAV Images with Deep learning 24

Efficient 3D Reconstruction of Multiple Plants from UAV Imag...

引用

16th international conference on machine learning and computing (icmlc)

作者： Huang, Hong Wang, Zhuowei Zhao, Genping Guangdong Univ Technol Sch Comp Sci & Technol Guangzhou Guangdong Peoples R China

ISBN: (纸本)9798400709234

Acquiring the 3D structure of plants is a critical task in the agricultural industry. Existing methods of generating 3D point clouds for multiple plants require a long processing time. In this paper, a 3D reconstruction method for numerous plants is proposed. Firstly, camera parameters in different viewpoints are obtained from the aerial image of plants by incremental structure from motion. Subsequently, the learning-based multi-view stereo takes images and the corresponding camera parameters as inputs to acquire initial depth maps. Finally, the depth maps are filtered and fused to produce a complete and dense 3D point cloud. We conducted experiments on an agricultural orchard dataset to compare with other methods. Experimental results demonstrate that our method reconstructs point clouds of plants with good quality while having a lower running time.

关键词： 3D reconstruction Plant Point cloud UAV images Deep learning

来源：评论

学校读者我要写书评

暂无评论

Visual-Semantic Alignment for Few-shot Remote Sensing Scene Classification 24

Visual-Semantic Alignment for Few-shot Remote Sensing Scene ...

引用

16th international conference on machine learning and computing (icmlc)

作者： Li, Haojun Li, Linjia Luo, Wei South China Agr Univ Pazhou Lab Guangzhou Peoples R China

ISBN: (纸本)9798400709234

We propose a few-shot learning approach that aligns visual and semantic features in an embedding feature space to alleviate the shortage of training (or reference) data in remote sensing scene classification (RSSC). Specifically, the self-supervised learning is first employed to improve the expressive ability of the learned feature, which could effectively enhance the features' generalizability. Meanwhile, we align the image feature and its corresponding class-semantic feature, which is obtained by feeding the class name to a language model such as BERT, to increase the image feature's discriminability. By systematically integrating the self-supervised learning and visual-semantic alignment with the backbone network, our approach could achieve image features with good generalizability and discriminability. Experiments on UCMerced LandUse, NWPU-RESISC45, and AID benchmarks validate the feasibility of our approach and verify its improved few-shot classification performance in RSSC.

关键词： Remote sensing scene classification Few-shot learning Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Character Expressions in Meta-learning for Extremely Low Resource Language Speech Recognition 24

Character Expressions in Meta-Learning for Extremely Low Res...

引用

16th international conference on machine learning and computing (icmlc)

作者： Zhou, Rui Ito, Akinori Nose, Takashi Tohoku Univ Grad Sch Engn Sendai Miyagi Japan

ISBN: (纸本)9798400709234

For the construction of a high-quality speech recognition system, a substantial volume of annotated speech data is requisite. However, preparing such expansive datasets is impracticable for a vast majority of global languages. Therefore, we need to develop speech recognition systems for low-resource languages. In this paper, we propose a method where the model is initially pretrained on speech data from ten languages utilizing meta-learning. After this, the model undergoes fine-tuning using a small amount of speech data of the target language. We used only about 15 minutes of speech and acheved CER of less 30%. Although the set of alphabet differs from language to language, alphabets often show phonetic resemblances. Capitalizing on this observation, we proposed the method of alphabet unification, employed English pronunciation rules based on Latin alphabet as a standard to align the alphabetic representation across languages under study. Our results show that such alphabet unification approach enhanced the performance.

关键词： Low resource Meta-learning Speech recognition Character unification

来源：评论

学校读者我要写书评

暂无评论

Multi-task learning Model for Location Estimation Using RSSI of Wireless LAN in NLoS Environment 24

Multi-task Learning Model for Location Estimation Using RSSI...

引用

16th international conference on machine learning and computing (icmlc)

作者： Wei, Zehui Ding, Jianbing Wang, Zening Wang, Xidong Ye, Xiaozhou Ouyang, Ye AsiaInfo Technol China Inc Beijing Peoples R China

ISBN: (纸本)9798400709234

In 6G networks, applying native AI/ML techniques to user signal quality data to obtain high-precision location estimation is a typical application scenario. Recent advanced techniques using global positioning system (GPS) and traditional received signal strength-based localization approaches are model-based. However, their accuracy decreases under the influence of non-line-of-sight (NLoS) reception and multipath fading channel, which leads to demand for data-oriented localization techniques using received signal strength indicator (RSSI) obtained from the surrounding radio transmitters with the help of powerful AI/ML techniques. AI/ML-based localization technique is quite challenging in practice for data scarcity issues and how to capture and fully utilize the information for records. In this work, we conducted careful feature engineering and proposed a multi-task learning model to cope with these problems. The multi-task learning model improves the estimation through explicitly extracting the RSSI distance relation with shared layer and coping with latitude and longitude simultaneously. We chose three tree-based models as baselines. Numerical experiment shows that the multi-task learning model is promising in the RSSI-based location estimation.

关键词： location estimation machine learning multi-task learning NLOS RSSI

来源：评论

学校读者我要写书评

暂无评论

Soft Adversarial Offline Reinforcement learning via Reducing the Attack Strength for Generalization 24

Soft Adversarial Offline Reinforcement Learning via Reducing...

引用

16th international conference on machine learning and computing (icmlc)

作者： Qiao, Wandi Yang, Rui Univ Sci & Technol China Hefei Peoples R China

ISBN: (纸本)9798400709234

Improving the generalization ability in offline reinforcement learning (RL) has received much attention in recent years. Existing adversarial RL approaches use adversarial training for the policy improvement, thus enhancing the generalization ability of RL agents. However, adversarial training severely hinders the performance improvement of agents in offline RL settings. This is because adversarial training is a pessimistic learning paradigm, where the adversarial attack patterns aim to improve the agents' generalization ability in worst-case scenarios. Such a learning paradigm struggles to improve the policy performance in the unstable training process of offline RL, thereby making it challenging to enhance generalization ability. To tackle this problem, we propose a novel offline adversarial RL approach, namely Soft Adversarial Offline Reinforcement learning (SAORL), which learns soft adversarial examples by reducing the attack strength of adversarial examples in offline RL. Specifically, SAORL proposes the Wasserstein-based constraint on traditional adversarial examples, thus formulating a worse-case optimization problem to learn the soft adversarial examples. We conduct extensive experiments on D4RL to evaluate our approach, which demonstrates SAORL can improve agents' performance and zero-shot generalization ability.

关键词： Offline Reinforcement learning Adversarial Training Zero-Shot Dynamics Generalization

来源：评论

学校读者我要写书评

暂无评论

learning Presolver Selection for Mixed-Integer Linear Programming 24

Learning Presolver Selection for Mixed-Integer Linear Progra...

引用

16th international conference on machine learning and computing (icmlc)

作者： Song, Wentao Gu, Naijie Univ Sci & Technol China Hefei Peoples R China

ISBN: (纸本)9798400709234

Presolving has become an important component of modern Mixed-integer linear programming (MILP) solvers. Empirically, it has been observed that the performance of the solver is significantly influenced by the presolving algorithm (presolver), and selecting an appropriate combination of presolvers can improve solving efficiency. In industry, it is common to manually control the switch of presolvers to find a more efficient combination. However, with the emergence of new presolvers, it has become increasingly challenging to manually implement an optimal selection strategy in the vast combination space. Therefore, this paper proposes presolver selection, which needs to consider two key issues: (P1) How many presolvers should be selected? (P2) Which presolvers should be preferred among them? To address this challenge, this paper uses a hierarchical sequence model (HEM) to learn the presolver selection strategy through reinforcement learning. Specifically, the high-level model learns how many presolvers should be selected, and the low-level model learns to select a subset of presolvers within the determined size. The experimental results show that the method used can solve (P1) and (P2) better scompared to the designed baseline. It effectively improves the performance of the solver on real-world and synthetic MILPs.

关键词： Mixed-integer Linear Programming Presolver Selection Hierarchical Sequence Model Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Classification and Segmentation of Apple Diseases Based on Deep learning 24

Classification and Segmentation of Apple Diseases Based on D...

引用

16th international conference on machine learning and computing (icmlc)

作者： Huang, Huanzhou Wang, Zhuowei Zhao, Genping Guangdong Univ Technol Sch Comp Sci & Technol Guangzhou Peoples R China

ISBN: (纸本)9798400709234

Aiming to address the issue of low accuracy in existing algorithms due to the limited scale of specific apple disease datasets and complex background information, a deep learning method integrating apple disease classification and segmentation is proposed. Initially, the method employs the CycleGAN network for data enhancement to combat overfitting problems in deep learning. Secondly, it constructs a deep residual disease classification network embedded in the convolutional block attention module. In essence, the ResNet50(*) module extracts fine-grained visual features and incorporates both the channel attention mechanism and spatial attention mechanism to focus on two critical features, thereby enhancing the disease classification performance of the model across various scenarios. Lastly, a hierarchical semantic segmentation model is introduced to achieve a more accurate segmentation effect through segmented segmentation. Experimental results demonstrate that the proposed method achieves an accuracy of 97.82% on the apple disease classification task, while obtaining MPA and MIoU scores of 94.85% and 91.75%, respectively, on the segmentation task.

关键词： Apple disease classification and segmentation Deep learning Convolutional block attention module Deep residual network Hierarchical semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Lightweight Object Detection-Tracking using Deep Feature Distillation 24

Lightweight Object Detection-Tracking using Deep Feature Dis...

引用

16th international conference on machine learning and computing (icmlc)

作者： Xue, B. Zheng, Q. Li, Z. Zhao, W. Wang, H. Feng, X. Tsinghua Univ Xian Jiaotong Univ Natl Univ Def Technol Beijing Peoples R China Tongji Univ Xian Jiaotong Univ Shanghai Peoples R China Natl Univ Def Technol Changsha Peoples R China Tsinghua Univ Beijing Peoples R China

ISBN: (纸本)9798400709234

Object detection and tracking are critical and fundamental problems in machine vision task. In this paper, an object detection and tracking method is proposed based on deep feature distillation. Particularly, an adaptive unsupervised Teacher-Student unified framework is developed. The Teacher module is performed by an expandable generative adversarial network mixture model. And knowledge discrepancy ranking (KDR) is designed to optimize Teacher resource allocation with the historical underlying knowledge. The Student module is developed based on a lightweight probabilistic generative model. And an unsupervised learning scheme is presented based on Gumbel-Soft sampling optimization to train jointly. A series of experiments are performed on authoritative dataset, demonstrating that the proposed method outperforms the state-of-the-art comparison methods.

关键词： Object detection object tracking deep learning knowledge distillation

来源：评论

学校读者我要写书评

暂无评论

FedRL: Federated learning with Non-IID Data via Review learning 24

FedRL: Federated Learning with Non-IID Data via Review Learn...

引用

16th international conference on machine learning and computing (icmlc)

作者： Wang, Jinbo Wang, Ruijin Pei, Xikai Univ Elect Sci & Technol China Chengdu Peoples R China

ISBN: (纸本)9798400709234

Federated learning epitomizes a sophisticated distributed machine learning methodology, enabling collaborative neural network model training across multiple entities without necessitating the transfer of local data, thereby fortifying data privacy protection. A significant challenge in federated learning lies in the statistical heterogeneity, characterized by non-independent and identically distributed (Non-IID) local data across diverse parties. This heterogeneity can engender inconsistent optimization within individual local models. Although previous research has endeavored to tackle issues stemming from heterogeneous data, our findings indicate that these attempts have not yielded high-performance neural network models. To confront this fundamental challenge, we introduce the FedRL framework in this paper, which facilitates efficient federated learning through review learning. The core principle of FedRL involves leveraging the knowledge representation generated by the global and local model layers to conduct periodic layer-by-layer comparative learning in a reciprocal manner. This strategy rectifies local model training, leading to enhanced outcomes. Our experimental results and subsequent analysis substantiate that FedRL effectively augments model accuracy in image classification tasks, while demonstrating resilience to statistical heterogeneity across all participating entities.

关键词： machine learning Federated learning Heterogeneity data Image classification

来源：评论

学校读者我要写书评

暂无评论

Continual Few-Shot Relation learning Via Dynamic Margin Loss and Space Recall 24

Continual Few-Shot Relation Learning Via Dynamic Margin Loss...

引用

16th international conference on machine learning and computing (icmlc)

作者： Li, Yongbing Duan, Pengfei Rong, Yi Yang, Yiwen Wang, Aoxing Wuhan Univ Technol Sch Comp Sci & Artificial Intelligence Wuhan Peoples R China Wuhan Univ Technol Sanya Sci & Educ Innovat Pk Sanya Peoples R China

ISBN: (纸本)9798400709234

Different from conventional continual relation learning (CRL), in continual few-shot relation learning (CFRL), there are only a small number of training samples for learning novel unseen relations. However, existing approaches using external memory for CFRL tasks only select one key instance for each relation and store it in memory which causes the feature space distribution characteristics of the old relations to be forgotten. In this paper, we propose a novel memory-based CFRL method to solve the above catastrophic forgetting problem. Specifically, we design a Dynamic Margin Loss to use the overall cosine similarity stored in memory to constrain the similarity between different relations to obtain a more ideal feature space distribution. Moreover, to let the model remember the overall distribution feature information of the seen relations, we propose a Space Recall constraint to Make the model re-member the overall feature space distribution range of the old relationship by learning the L2 distance stored in memory. Extensive experiments demonstrate that the proposed method outperforms previous state-of-the-art methods by 0.75%, 1.74% and 1.34% in the 10-way-2-shot, 10-way-5-shot and 10way-10-shot settings.

关键词： Continual learning Few-shot learning Relation Extraction

来源：评论

学校读者我要写书评

暂无评论

没有更多数据了...

全选清除本页清除全部题录导出标记到“检索档案”

共195页 << < 1 2 3 4 5 6 7 8 9 10 > >>

检索报告对象比较合并检索0

隐藏清空

合并搜索

回到顶部

执行限定条件

内容：

评分：

请选择保存的检索档案：

请选择收藏分类：

订阅名称：

通借通还

温馨提示：

图书名称：

借书校区：

取书校区：

手机号码：

邮箱地址：

一卡通帐号：

电话和邮箱必须正确填写，我们会与您联系确认。

联系人：

所在院系：

联系邮箱：

联系电话：

内蒙古自治区呼和浩特市赛罕区大学西街235号邮编: 010021

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：