生成式人工智能赋能的非线性学习智能体模型建构  

Construction of a Nonlinear Learning Agent Model Empowered by Generative Artificial Intelligence

作　　者：黄红涛 余琳 王继新[2] Huang Hongtao;Yu Lin;Wang Jixin(Office of Informatization,Central China Normal University,Wuhan 430079,Hubei;School of Artificial Intelligence in Education,Central China Normal University,Wuhan 430079,Hubei)

机构地区：[1]华中师范大学信息化办公室,湖北武汉430079 [2]华中师范大学人工智能教育学部,湖北武汉430079

出　　处：《中国电化教育》2024年第11期61-68,共8页China Educational Technology

基　　金：国家社会科学基金教育学国家重点项目“中国数字教育实践的理论建构研究”(项目批准号:ACA240018)研究成果。

摘　　要：该研究探讨了生成式人工智能(AIGC)赋能的非线性学习智能体模型建构,分析了非线性学习的特点,并预设了自学、混合和协作三种学习场境。基于这些场境推导出智能体模型的功能需求,进而抽象出模型的核心能力。该文界定了非线性学习智能体的含义、数学表达式及核心算法,并阐述智能体模型的核心是一个改进的马尔可夫决策过程,旨在能够为学习者提供个性化、动态优化的学习支持,有效提升非线性学习的效率,为AIGC在教育领域的应用提供了新的模型构建方法,对推动教育信息化和个性化学习具有重要意义。This study explores the construction of a nonlinear learning agent model empowered by generative artificial intelligence(AIGC).The research first analyzes the characteristics of nonlinear learning and presets three learning scenarios:self-learning,hybrid learning,and collaborative learning.Based on these scenarios,the functional requirements of the agent model are derived,and the core capabilities of the model are abstracted.The paper defines the concept,mathematical expression,and core algorithms of the nonlinear learning agent,emphasizing that the core of the agent model is an improved Markov decision process.The study aims to provide personalized,dynamically optimized learning support to learners,effectively enhancing the efficiency of nonlinear learning.This research offers a novel approach to model construction for AIGC applications in education,holding significant implications for advancing educational informatization and personalized learning.

关 键 词：生成式人工智能 非线性学习 智能体模型 个性化学习 教育信息化 

分 类 号：G434[文化科学—教育学]