It is our great pleasure to announce the publication of this special section in JCST:"Self-Learning with Deep Neural Networks". Self-learning is an important skill for human beings as they journey through education and beyond to advisors,building independence and ability to progress without reliance on a teacher.Recently,as a crucial branch of artificial intelligence,self-learning with deep neural networks sheds its light on diverse research directions,e.g.,self-supervised learning,self-distillation learning,self-attention learning,and adversarial learning.Also,excellent results have been achieved in many application tasks in computer vision and natural language processing by leveraging these self-learning approaches.Therefore,for better understanding and developing self-learning methods,it is desirable to conduct in-depth research on self-learning with deep neural networks from both theoretical and applied perspectives.

Min-Ling Zhang;Xiu-Shen Wei;Gao Huang

School of Computer Science and Engineering,Southeast University,Nanjing;School of Computer Science and Engineering,Nanjing University of Science and Technology,Nanjing;Associate Professor,Department of Automation,Tsinghua University,Beijing

计算机科学技术学报（英文版）

[1]Min-Ling Zhang;Xiu-Shen Wei;Gao Huang-.Preface)[J].计算机科学技术学报（英文版）,2022(03):505-506

JCST, Self,advisors

Preface,It,great,pleasure,announce,publication,this,special,section,Learning,Deep,Neural,Networks,learning,important,skill,human,beings,they,journey,through,education,beyond,building,independence,ability,progress,without,reliance,teacher,Recently,crucial,branch,artificial,intelligence,self,deep,neural,networks,sheds,its,light,diverse,research,directions,supervised,distillation,attention,adversarial,Also,excellent,results,have,been,achieved,many,application,tasks,computer,vision,natural,language,processing,by,leveraging,these,approaches,Therefore,better,understanding,developing,methods,desirable,conduct,depth,from,both,theoretical,applied,perspectives

0.672525