检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Zhang, Chunxu Cui, Jiaxu Yang, Bo College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

In recent years, deep learning has achieved remarkable achievements in many fields, including computer vision, natural language processing, speech recognition and others. Adequate training data is the key to ensure the effectiveness of the deep models. However, obtaining valid data requires a lot of time and labor resources. Data augmentation (DA) is an effective alternative, which can generate new labeled data based on existing data using label-preserving transformations. Although we can benefit a lot from DA, designing appropriate DA policies requires a lot of expert experience and time consumption, and the evaluation of searching the optimal policies is costly. So we raise a new question in this paper: how to achieve automated data augmentation at as low cost as possible? We propose a method named BO-Aug for automating the process by finding the optimal DA policies using the Bayesian optimization approach. We validate the BO-Aug on three widely used image classification datasets, including CIFAR-10, CIFAR- 100 and SVHN. Experimental results show that the proposed method can achieve state-of-the-art or near advanced classification accuracy at a relatively low policies search cost, and the searched policies based on a specific dataset are transferable across different neural network architectures or even different datasets. Code to reproduce our experiments is available at https://***/zhangxiaozao/BOAug. Copyright © 2019, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Explanation Chains Model Based on the Fine-Grained Data 8th

Explanation Chains Model Based on the Fine-Grained Data

引用

8th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2019

作者： Ma, Fu-Yuan Chen, Wen-Qi Xiao, Min-Hao Wang, Xin Wang, Ying College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbol Computation and Knowledge Engineering Jilin University Ministry of Education Changchun130012 China Changchun Institute of Technology Changchun130012 China College of Software Jilin University Changchun130012 China

ISBN: (纸本)9783030322359

With the development of information society, Recommendation System has been an import tool to help users filter information and create more economic value for enterprises. However, it is difficult for traditional recommendation systems to interpret recommendation results. In order to improve users’ trust in recommendation results, interpretable recommendation models have attracted more and more attention. In this paper, we present the Explanation Chains Model based on the Fine-grained Data (F-ECM) to enhance the effectiveness of recommendation while achieving the interpretability of recommendation. First, we generate parsing trees from user comments and extract three key sentence structure information (i.e., aspects, features and sentiment tendency) from those generated parsing tree. The fine-grained similarity is computed based on the aspects and features of the products to be recommended, and users’ product satisfaction is predicted by combining sentiment tendency. Then the recommendation chain will be constructed according to the satisfaction degree in the recommendation list. Finally, we calculate recommendation chain scores of all the items to be recommended to the target user, generate the recommendation results and personal explanation for the user by the recommendation chains. Experiments in the Amazon data set show that the Explanation Chains Model based on the Fine-grained Data achieve better interpretability and performance of product recommendation systems. © 2019, Springer Nature Switzerland AG.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Recruitment-imitation mechanism for evolutionary reinforcement learning

arXiv

引用

arXiv 2019年

作者： Lü, Shuai Han, Shuai Zhou, Wenbo Zhang, Junwei Key Laboratory of Symbolic Computation and Knowledge Engineering Jilin University Ministry of Education Changchun130012 China College of Computer Science and Technology Jilin University Changchun130012 China

Reinforcement learning, evolutionary algorithms and imitation learning are three principal methods to deal with continuous control tasks. Reinforcement learning is sample efficient, yet sensitive to hyper-parameters setting and needs efficient exploration;Evolutionary algorithms are stable, but with low sample efficiency;Imitation learning is both sample efficient and stable, however it requires the guidance of expert data. In this paper, we propose Recruitment-imitation Mechanism (RIM) for evolutionary reinforcement learning, a scalable framework that combines advantages of the three methods mentioned above. The core of this framework is a dual-actors and single critic reinforcement learning agent. This agent can recruit high-fitness actors from the population of evolutionary algorithms, which instructs itself to learn from experience replay buffer. At the same time, low-fitness actors in the evolutionary population can imitate behavior patterns of the reinforcement learning agent and improve their adaptability. Reinforcement and imitation learners in this framework can be replaced with any off-policy actor-critic reinforcement learner or data-driven imitation learner. We evaluate RIM on a series of benchmarks for continuous control tasks in Mujoco. The experimental results show that RIM outperforms prior evolutionary or reinforcement learning methods. The performance of RIM's components is significantly better than components of previous evolutionary reinforcement learning algorithm, and the recruitment using soft update enables reinforcement learning agent to learn faster than that using hard update. Copyright © 2019, The Authors. All rights reserved.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

A pseudo label based dataless naive bayes algorithm for text classification with seed words 27

A pseudo label based dataless naive bayes algorithm for text...

引用

27th International Conference on computational Linguistics, COLING 2018

作者： Li, Ximing Yang, Bo College of Computer Science and Technology Jilin University China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education China

ISBN: (纸本)9781948087506

Traditional supervised text classifiers require a large number of manually labeled documents, which are often expensive to obtain. Recently, dataless text classification has attracted more attention, since it only requires very few seed words of categories that are much cheaper. In this paper, we develop a pseudo-label based dataless Naive Bayes (PL-DNB) classifier with seed words. We initialize pseudo-labels for each document using seed word occurrences, and employ the expectation maximization algorithm to train PL-DNB in a semi-supervised manner. The pseudo-labels are iteratively updated using a mixture of seed word occurrences and estimations of label posteriors. To avoid noisy pseudo-labels, we also consider the information of nearest neighboring documents in the pseudo-label update step, i.e., preserving local neighborhood structure of documents. We empirically show that PL-DNB outperforms traditional dataless text classification algorithms with seed words. Especially, PL-DNB performs well on the imbalanced dataset. © 2018 COLING 2018 - 27th International Conference on computational Linguistics, Proceedings. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

A Registration Method for Augmented Reality System Based on Visual SLAM

A Registration Method for Augmented Reality System Based on ...

引用

International Conference on Electronic engineering and Informatics (EEI)

作者： Zuoming Zhang Man Shu Zixuan Wang Hanwen Wang Xin Wang College of Software Engineering Jilin University Changchun China College of Computer Science and Technology Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineer of Ministry of Education Jilin University Changchun China

ISBN: (纸本)9781728140773

In augmented reality, smart devices need to sense their own position in real physical space and complex scene structure to achieve good virtual reality interaction and three-dimensional registration. This paper proposes a technical framework of visual SLAM, which uses VO and BA methods based on feature points to obtain pose, and applies it to the augmented reality system to realize pose estimation, registration, tracking, collision and other effects of intelligent equipment independent of artificial or natural landmarks.

关键词： augmented reality image registration pose estimation robot vision SLAM (robots) augmented reality pose estimation robot vision image registration supersonic low altitude missile Registration

来源：评论

学校读者我要写书评

暂无评论

Black-box expectation propagation for Bayesian models

Black-box expectation propagation for Bayesian models

引用

2018 SIAM International Conference on Data Mining, SDM 2018

作者： Li, Ximing Li, Changchun Chi, Jinjin Ouyang, Jihong Wang, Wenting College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University China College of Mathematics Jilin University China

In this paper, we develop a generic black-box expectation propagation (BBEP) algorithm that can be directly applied to Bayesian models without model-specific derivations. BBEP is built on the spirit of using Monte Carlo estimates, where the moment matching step in EP is replaced with Monte Carlo approximations. To avoid high variance, we employ importance sampling for variance reduction and analyze how to find an optimal proposal distribution. We compare BBEP against the state-of-the-art black-box algorithms on both synthetic and real-world data sets. The experimental results indicate that BBEP can reach better predictive performance than baseline algorithms, and even can be on a par with analytical solutions in some settings. © 2018 by SIAM.

关键词： Importance sampling

来源：评论

学校读者我要写书评

暂无评论

Multiple Auxiliary Information Based Deep Model for Collaborative Filtering

引用

Journal of Computer Science & Technology 2018年第4期33卷 668-681页

作者： Lin Yue Xiao-Xin Sun Wen-Zhu Gao Guo-Zhong Feng Bang-Zuo Zhang School of Information Science and Technology Northeast Normal University Changchun 130117 China Key Laboratory of Applied Statistics of Ministry of Education Northeast Normal University Changchun 130024 China School of Environment Northeast Normal University Changchun 130117 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

With the ever-growing dynamicity, complexity, technique is proposed and becomes one of the most effective and volume of information resources, the recommendation techniques for solving the so-called problem of information overload. Traditional recommendation algorithms, such as collaborative filtering based on the user or item, only measure the degree of similarity between users or items with single criterion, i.e., ratings. According to the experience of previous studies, single criterion cannot accurately measure the similarity between user preferences or items. In recent years, the application of deep learning techniques has gained significant momentum in recommender systems for better understanding of user preferences, item characteristics, and historical interactions. In this work, we integrate plot information as auxiliary information into the denoising autoencoder （DAE）, called SemRe-DCF, which aims at learning semantic representations of item descriptions and succeeds in capturing fine-grained semantic regularities by using vector arithmetic to get better rating prediction. The results manifest that the proposed method can effectively improve the accuracy of prediction and solve the cold start problem.

关键词： semantic representation plot information denoising autoencoder collaborative filtering auxiliary information

来源：评论

学校读者我要写书评

暂无评论

Game Theoretic Approaches in Vehicular Networks: A Survey

arXiv

引用

arXiv 2020年

作者： Sun, Zemin Liu, Yanheng Wang, Jian Anil, Carie Cao, Dongpu College of Computer Science and Technology Key Laboratory of Symbolic Computation Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China Engineering VIT-AP AmaravatiAndhra Pradesh522237 India College of Engineering Nanjing Agricultural University Nanjing210031 China Waterloo Cognitive Autonomous Driving Lab University of Waterloo N2L 3G1 Canada

—In the era of the Internet of Things (IoT), vehicles and other intelligent components in Intelligent Transportation System (ITS) are connected, forming the Vehicular Networks (VNs) that provide efficient and secure traffic, ubiquitous access to information, and various applications. However, as the number of connected nodes keeps increasing, it is challenging to satisfy various and large amounts of service requests with different Quality of Service (QoS) and security requirements in the highly dynamic VNs. Intelligent nodes in VNs can compete or cooperate for limited network resources so that either an individual or a group objectives can be achieved. Game theory, a theoretical framework designed for strategic interactions among rational decision-makers who faced with scarce resources, can be used to model and analyze individual or group behaviors of communication entities in VNs. This paper primarily surveys the recent advantages of GT used in solving various challenges in VNs. As VNs and GT have been extensively investigate34d, this survey starts with a brief introduction of the basic concept and classification of GT used in VNs. Then, a comprehensive review of applications of GT in VNs is presented, which primarily covers the aspects of QoS and security. Moreover, with the development of fifth-generation (5G) wireless communication, recent contributions of GT to diverse emerging technologies of 5G integrated into VNs are surveyed in this paper. Finally, several key research challenges and possible solutions for applying GT in VNs are outlined. Copyright © 2020, The Authors. All rights reserved.

关键词： Game theory

来源：评论

学校读者我要写书评

暂无评论

SNB-PPB: Social-network-based-privacy-preserving Broadcast for Vehicular Communications 4

SNB-PPB: Social-network-based-privacy-preserving Broadcast f...

引用

4th International Conference on Vehicle Technology and Intelligent Transport Systems, VEHITS 2018

作者： Liu, Yanheng Zhu, Haifeng Wang, Jian College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China

ISBN: (纸本)9789897582936

Internet key Exchange (IKE) can be used in vehicular communications in vehicular ad hoc networks (VANETs), but it requires a stable communication procedure, which is usually difficult to achieve because of the highly dynamic context of a VANET. In this paper, we propose a new method named social-network-based-privacy-preserving broadcast (SNB-PPB) for secure vehicular communications, which can omit the key exchange procedure. It transfers the information of online social networks to offline vehicular communications, dividing messages into different privacy levels and vehicles into different trust levels based on trust relations on social networks. The sender uses attributes on social networks and the message’s privacy level to encrypt messages;only those receivers who satisfy the corresponding trust level can decrypt the broadcast *** the best of our knowledge, this is the first time that social networks have been applied in VANETs to achieve secure vehicular communications and privacy preservation. Our simulation results show that our method enables the sender to decide the amount and percentage of vehicles that can decrypt the received broadcast. Copyright © 2018 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

A Method of Identifying Thunderstorm Clouds in Satellite Cloud Image Based on Clustering

引用

Computers, Materials & Continua 2018年第12期57卷 549-570页

作者： Lili He Dantong Ouyang Meng Wang Hongtao Bai Qianlong Yang Yaqing Liu Yu Jiang College of Computer Science and Technology Jilin UniversityChangchun130012China Key Laboratory of Symbolic Computation and Knowledge Engineering Jilin UniversityChangchun130012China School of Information Science&Technology Dalian Maritime UniversityDalian116026China School of Informatics University of EdinburghEdinburghEH89ABUK

In this paper,the clustering analysis is applied to the satellite image segmentation,and a cloud-based thunderstorm cloud recognition method is proposed in combination with the strong cloud computing *** method firstly adopts the fuzzy C-means clustering(FCM)to obtain the satellite cloud image ***,in the cloud image,we dispose the‘high-density connected’pixels in the same cloud clusters and the‘low-density connected’pixels in different cloud ***,we apply the DBSCAN algorithm to the cloud image obtained in the first step to realize cloud cluster ***,using the method of spectral threshold recognition and texture feature recognition in the steps of cloud clusters,thunderstorm cloud clusters are quickly and accurately *** experimental results show that cluster analysis has high research and application value in the segmentation processing of meteorological satellite cloud images.

关键词： Cloud computing cluster analysis FCM DBSCAN thunderstorm clouds satellite cloud image

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：