基于项目驱动与多维融合的研究生机器学习课程教学改革与实践

Teaching Reform and Practice of Graduate-Level Machine Learning Course Based on Project-Driven and Multi-Dimensional Integration

摘要: 针对人工智能时代机器学习课程教学中存在的理论教学与实践应用脱节、学生创新能力培养不足等问题，本文提出一种基于项目驱动与多维融合的课程教学改革方案。该方案以学生为中心，构建“前瞻基础–核心知识–扩展创新”三层次课程内容体系，创新“案例式 + 项目式 + 研讨式”多元融合教学模式，并建立以过程为导向的形成性评价机制。通过连续两年的教学实践对比分析表明，改革实施后，学生在实践创新能力和综合应用能力等方面均有显著提升。该改革实践为研究生机器学习课程建设提供了可借鉴的经验。

Abstract: Addressing the issues of disconnection between theoretical teaching and practical application, as well as the insufficient cultivation of students’ innovative capabilities in machine learning courses in the era of artificial intelligence, this paper proposes a curriculum reform scheme based on project-driven and multi-dimensional integration. This student-centered scheme constructs a three-level course content system comprising forward foundation, core knowledge and extended innovation, innovates a multi-dimensional integrated teaching model combining “Case-based”, “Project-based” “Seminar-based” approaches, and establishes a process-oriented formative assessment mechanism. Through a comparative analysis of two consecutive years of teaching practice, the results show that students’ practical innovation capabilities and comprehensive application abilities have been significantly improved. This reform practice provides a valuable reference for the development of graduate-level machine learning courses.