期刊文献+

从机器学习走向人机共学:以语言为桥梁

From Machine Learning to Human-AI Collaborative Learning:Bridging through Language
下载PDF
导出
摘要 语言提供了超越对话的审视机器学习与人类学习的更高维度,是生成式人工智能时代从机器学习走向人机共学的桥梁。基于已有文献,本研究从语言视域比较人类学习和机器学习的同一性与差异性,以把握二者的本质。本研究认为,人类学习与其镜像机器学习之同一性表现为二者均是对我与世界的解码与编码,是我与我、我与他者的交互,是创建知识链接与脑的重塑,是多模态转换的输入与输出。人机学习的差异性在于人类“为己之学”对机器“为人之学”的超越、人类“大写之学”对机器“小写之学”的超越、人类“具身学习”对机器“离身学习”的超越、人类“德性学习”对机器“无德学习”的超越。鉴于此,本研究提出建构应然样态的人机学习未来:从语言生活中人机共存到人机共学、从我与他者的人机对话到人机共智、从人类文明的语言规约到人机共善。本研究可为人类学习与机器学习提供语言视域下的新理解与新阐释,对推动机器与人类的共同发展与进化、构建和谐的人机关系具有启示意义。 Language offers a perspective that transcends beyond dialogue and provides a higherdimensional lens to examine machine and human learning.In the era of generative artificial intelligence,language serves as a bridge from machine learning to the emerging paradigm of humanmachine co-learning.Based on literature and a linguistic framework,this study compares the similarities and differences between human learning and machine learning,aiming to grasp the essence of both through the commonalities and distinctions.This study posits that the core similarity between human and machine learning lies in their shared processes of decoding and encoding the relationship between the self and the world.Both involve interactions between the self and others,creating knowledge networks,restructuring the brain,and achieving multimodal input-output transformations.However,their differences include human learning,in the form of“capital-L Learning,”which is“learning for self,”and machine learning,in“lowercase-L learning,”which is fundamentally“learning for others.”Moreover,embodied human learning stands in contrast to the disembodied nature of machine learning,and human“ethical learning”transcends the“ethics-free learning”of machines.In light of these differences,the study advocates for constructing a future where humanmachine co-learning reaches its ideal state.This includes envisioning humans and machines coexist in shared linguistic environments,progressing from human-machine dialogues to human-machine collective intelligence,and transitioning from the linguistic norms of human civilization to a future of human-machine co-flourishing.This research offers a new understanding and interpretation of human and machine learning from a linguistic perspective,providing practical insights for advancing the mutual development and evolution of humans and machines and building a harmonious human-machine relationship.
作者 李永梅 谭维智 LI Yongmei;TAN Weizhi(Faculty of Education,Qufu Normal University,Qufu 273165,China)
出处 《开放教育研究》 CSSCI 北大核心 2024年第6期4-12,共9页 Open Education Research
基金 2024年度教育部人文社会科学研究青年基金项目“知识生产模式转型视角下教师数字素养提升研究”(24YJC880102) 山东省社科联2023年度人文社会科学课题“AIGC时代教育的知识生产与传播创新研究”(2023-JCXK-012)。
关键词 人类学习 机器学习 语言 生成式人工智能 人机共学 human learning machine learning language generative artificial intelligence humanmachine co-learning
  • 相关文献

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部