摘要
教育心理学研究的主要目的是了解人是如何学习的,人工智能神经网络研究的核心在于探究机器是如何学习的。教育心理学经过一个多世纪的发展,有诸多成熟的方法和理论。ICAP学习方式分类学,是国际教育心理学领域最新取得的一个重大研究成果,其研究方法和结论对人工智能神经网络设计有什么价值?本文假设ICAP四种学习方式与人工智能神经网络设计元素之间能建立起一种关联,将目前流行的人工智能神经网络构成元素和算法依据ICAP的四种学习方式进行分类拆分,以一个基本教学设计为案例,依据学习方式对应的人工智能神经网络构成元素设计了简单的人工智能神经网络。文章尝试在人类学习与机器学习之间建立某种对应参照关系,为人工智能神经网络设计研究提供一个新思路。
The main purpose of educational psychology research is to understand how knowledge is learned. The core of artificial intelligence neural network research is to explore how the machine learns. The development of educational psychology over more than a century has produced plenty of mature theories and methods. The ICAP Framework is a major achievement in the field of educational psychology. How do the research methods and conclusion contribute to the artificial intelligence neural network design7 In this article, we assume that there is a relationship between the four modes of ICAP and the design elements of artificial intelligence network. Components and algorithms of the current artificial intelligence neural network are classified into groups in accordance with the four modes of ICAP. On the basis of the principles of instructional design, a simple artificial intelligence network design is carried out according to the components of the artificial intelligence neural network classified in line with the modes of ICAP. This paper attempts to establish a corresponding reference relationship between human learning and machine learning while at the same time gives new insight into artificial intelligence neural network design research.
作者
熊媛
王铭军
盛群力
Xiong Yuan1,2, Wang Mingjun3, Sheng Qunli4(1 .School of Education, Tianjin University, Tianjin 300350; 2.School of International Studies, Zhejiang Business College Hangzhou Zhejiang 310053; 3. College of Engineering, Lishui University, Lishui Zhejiang 323000; 4. College of Education, Zhejiang University, Hangzhou Zhejiang 31002)
出处
《中国电化教育》
CSSCI
北大核心
2018年第11期118-125,共8页
China Educational Technology
关键词
学习方式分类学
人工智能神经网络
网络设计
机器学习
The ICAP Framework
Artificial Intelligence Neural Network
Network Design
Machine Learning
