专刊简介:机器学习

中国电子学会  |   2020-12-18 16:56

来源:电子学报英文

专刊简介

人工智能 (AI) 可能是迄今为止人类最神奇和复杂的创造,对工业发展乃至整个现代社会产生了革命性影响。人工智能的研究目的是让机器实现“类人”功能。依照系统可实现类人程度的高低,AI可分为弱人工智能、通用人工智能、超级人工智能。机器学习(ML)是弱人工智能的重要技术之一。

传统的机器学习方法,如树模型、聚类模型、贝叶斯模型、线性判别模型、人工神经网络和支持向量机等,被广泛研究和应用在机器视觉、自然语言理解、机器人、生物信息学、机器翻译、自动控制等领域中,取得了众多杰出成果。近年来,深度学习作为一种有代表性的表示学习方法脱颖而出。相较于传统的机器学习方法,基于深度学习的方法在上述领域中的识别性能有了显著提高。

本专刊着力于展示机器学习领域的创新工作,特别关注能够代表机器学习及AI前沿发展的观点和论述。在这些研究成果中,一方面介绍了基于进化学习、非均匀量化、改进的拉普拉斯矩阵、多层结构嵌入、量子互文性、流形学习等理论和方法,在深度学习理论及其应用中开展的创新工作。另一方面则侧重于不同类型的机器学习方法改进及其应用,其中包括四元核费舍尔判别分析、基于相对质量与二分空间树的森林方法、极限特征值方法、三阶马尔科夫链模型等。上述方法的应用领域涵盖了图像处理、安全和隐私、软件工程、云调度、无线通信、社交计算、AI辅助医疗分析、性能分析等。

广大研究者在机器学习领域的探索历久弥新且惊喜不断。在过去二十年的快速发展过程中,本领域的研究热点可以简要概括为:2000年至2006年的流形学习;2006年至2011年的稀疏学习,以及2012年至今的深度学习。机器学习的下一个热点将是什么,让我们翘首以待。


Introduction

Artificial intelligence (AI) is probably the most amazing and complex creation of mankind to date, which has the revolutionary impact on industrial development and even modern society. AI research tries to make machines realize human-like functioning. AI can be categorized into artificial narrow intelligence, artificial general intelligence, and artificial super intelligence based on the degree to which an AI system can achieve human capabilities. Machine learning (ML) is one of the important technologies in artificial narrow intelligence.

Traditional machine learning methods, including tree models, clustering models, Bayesian models, linear discriminant models, artificial neural networks, and support vector machines, well treat topics of research and applications in machine vision, natural language understanding, robots, bioinformatics, machine translation, automatic control, etc. In recent years, deep learning, an empirical method of representative learning, has improved the performance of above topics significantly.

This special issue is rich in the ideas of machine learning, and we try to introduce the ideas that have motivated the new progress of machine learning and AI. More than 15 papers implement research on the modification of deep learning methods and their applications. The optimization methods are based on evolutionary learning, non-uniform quantization, revised Laplacian matrix, hierarchical structure embedding, quantum contextuality, an expression from manifolds, respectively. Other papers focus on different kinds of machine learning methods and their applications, which consist of quaternion kernel Fisher discriminant analysis, relative mass and half-space tree based forest, extreme eigenvalues method, 3rd-order Markov chain model, etc. Applications broadly lie in image processing, security and privacy, software engineering, cloud scheduling, wireless communication, social computing, AI-aided medical analysis, performance analysis, etc.

Although it has been explored for a long time, machine learning still surprises us frequently. Throughout its development in the past two decades, research hotspots can be briefly summarized as manifold learning from 2000 to 2006, sparse learning from 2006 to 2011, and deep learning from 2012 to the present. Then, what will become the next research focus in machine learning, we can hardly wait to see.


本专刊为Chinese Journal of Electronics 2020年第6期。您可以在本文浏览本期目录。本期文章在本刊官网首页可供下载(点击文末“阅读原文”即可前往),在其他平台也已上线,您可以在 IET Digital Library、IEEE Xplore、知网查阅。

This special issue is available in Vol.29, No.6 of Chinese Journal of Electronics. Contents can be found in our official website by clicking "阅读原文" in the end of this page. It is also available on other online platforms (e.g., IET Digital Library, IEEE Xplore, CNKI).

CONTENTS

Vol.29 No.6                         Nov. 2020

PREFACE

991......New Progress in Research and Application of Machine Learning

SUN Guanglu

REVIEW

992......Deep Learning and Its Application in Diabetic Retinopathy Screening

ZOU Beiji, SHAN Xi, ZHU Chengzhang, DAI Yulan, YUE Kejuan, CHEN Yuanqiong, XIAO Yalong, HUANG Jiaer

1001.....Research and Application of Machine Learning in Automatic Program Generation

ZHANG Xiaojiang, JIANG Ying

ARTICLES

1016.....CSELM-QE: A Composite Semi-supervised Extreme Learning Machine with Unlabeled RSS Quality Estimation for Radio Map Construction

ZHAO Jianli, WANG Wei, SUN Qiuxia, HUO Huan, SUN Guoqiang, GAO Xiang, ZHU Chendi

1025.....Multi-level Deep Correlative Networks for Multi-modal Sentiment Analysis

CAI Guoyong, LYU Guangrui, LIN Yuming, WEN Yimin

1039.....Recurrent Neural Networks for Computing the Moore-Penrose Inverse with Momentum Learning

ZHANG Naimin, ZHANG Ting

1046.....Performance Analysis and Prediction of Double-Server Polling System Based on BP Neural Network

YANG Zhijun, MAO Lei, GAN Jianhou, DING Hongwei

1054......Malware Detection Algorithm Based on the Attention Mechanism and ResNet

WANG Lele, WANG Binqiang, ZHAO Peipei, LIU Ruyi, LIU Jiangang, MIAO Qiguang

1061......A Fast Evolutionary Learning to Optimize CNN

CHEN Jinyin, LIN Xiang, GAO Shangtengda, XIONG Hui, ZHANG Longyuan, LIU Yi, XUAN Qi

1074.....Image Inpainting Based on Improved Deep Convolutional Auto-encoder Network

QIANG Zhenping, HE Libo, DAI Fei, ZHANG Qinghui, LI Junqiu

1085.....Quaternion Kernel Fisher Discriminant Analysis for Feature-Level Multimodal Biometric Recognition

WANG Zhifang, ZHEN Jiaqi, ZHU Fuzhen, HAN Qi

1093.....RMHSForest: Relative Mass and Half-Space Tree Based Forest for Anomaly Detection

LYU Yanxia, LI Wenjie, WANG Yue, SUN Siqi, WANG Cuirong

1102.....DSCA-Net: Indoor Head Detection Network Using Dual-Stream Information and Channel Attention

QIN Pinle, SHEN Wenxiang, ZENG Jianchao

1110.....Student Performance Prediction Based on Behavior Process Similarity

BAO Yunxia, LU Faming, WANG Yanxiao, ZENG Qingtian, LIU Cong

1119.....Learning Domain-Invariant and Discriminative Features for Homogeneous Unsupervised Domain Adaptation

ZHANG Yun, WANG Nianbin, CAI Shaobin

1126.....CNQ: Compressor-Based Non-uniform Quantization of Deep Neural Networks

YUAN Yong, CHEN Chen, HU Xiyuan, PENG Silong

1134.....Advancing Graph Convolution Network with Revised Laplacian Matrix

WANG Jiahui, GUO Yi, WANG Zhihong, TANG Qifeng, WEN Xinxiu

1141.....HSNR: A Network Representation Learning Algorithm Using Hierarchical Structure Embedding

YE Zhonglin, ZHAO Haixing, ZHU Yu, XIAO Yuzhi

1153.....Feature Fusion Based Hand Gesture Recognition Method for Automotive Interfaces

XU Qianyi, QIN Guihe, SUN Minghui, YAN Jie, JIANG Huiming, ZHANG Zhonghan

1165.....Detection of Malicious PDF Files Using a Two-Stage Machine Learning Algorithm

HE Kang, ZHU Yuefei, HE Yubo, LIU Long, LU Bin, LIN Wei

1178.....Quantum Contextuality for Training Neural Networks

ZHANG Junwei, LI Zhao

1185.....The Kernel Dynamics of Convolutional Neural Networks in Manifolds

WU Wei, JING Xiaoyuan, DU Wencai

1193.....Classification and Early Warning Model of Terrorist Attacks Based on Optimal GCN

FENG Yong, GAI Ming, WANG Fuhai, WANG Rongbing, XU Xiaowei

1201.....A Novel Goodness of Fit Test Spectrum Sensing Using Extreme Eigenvalues

LI He, ZHAO Wenjing, LIU Chang, JIN Minglu, YOO Sang-Jo  

1207.....SLA-Aware and Energy-Efficient VM Consolidation in Cloud Data Centers Using Host State 3rd-Order Markov Chain Model

LI Lianpeng, DONG Jian, ZUO Decheng, JI Songyan

公众号:电子学报英文

微信号:CJEjournal

邮    箱:cje@ejournal.org.cn


点击下方“阅读原文”前往本刊官网浏览并下载文章

来源:CJEjournal 电子学报英文

原文链接:http://mp.weixin.qq.com/s?__biz=MzIzMzM2NTY2NQ==&mid=2247484345&idx=1&sn=a8c78d1952ae2bc93275f655442738d5

版权声明:除非特别注明,本站所载内容来源于互联网、微信公众号等公开渠道,不代表本站观点,仅供参考、交流、公益传播之目的。转载的稿件版权归原作者或机构所有,如有侵权,请联系删除。

电话:(010)86409582

邮箱:kejie@scimall.org.cn