上海交通大学研究生课程开设申请表 New Graduate Course Application Form, SJTU 课程基本信息 Basic Information *课程名称 Course Name (中文 Chinese) 材料信息学 – 材料中的大数据与机器学习 (英文 English) Materials Informatics – Big Data and Machine Learning in Materials *学分 Credits 3 *学时 Teaching Hours 48 (1 学分≥16 课时) *开课学期 Semester 秋季 *时否跨学期 Cross-semester? 否 跨 Spanning over 个 学 Fall 期 Semesters。 *课程性质 Course Category 专业课 *课程分类 Course Type 全日制 Major Course Full-time *授课语言 Instruction Language 中文 Chinese *成绩类型 Grade 等级制 Letter Grade *开课院系 School (050)材料科学与工程学院 School of Materials Science & Engineering 所属学科 Subject 0805 材料科学与工程 负责教师 Person in charge 姓名 Name 工号 ID 单位 School 联系方式 E-mail 汪洪 hongwang2@sjtu.edu.cn 课程扩展信息 Extended Information *课程简介 (中文) Course Description 近年来,大数据及人工智能作为新兴学科得到了快速发展和推广,在材料科学中体现 为材料信息学,成为今后材料科学与工程学科研究模式发展的趋势,有利于加快材料 的研发速度,降低成本。为适应这种模式转变,对接国家科技创新战略、顺应材料学 科发展潮流,特为研究生专门开设本课程。主要授课内容涉及数据驱动的材料科学概 述、数据库与数据标准、高通量计算、实验方法与案例,并将较为系统地讲授机器学 习算法及案例等。本课程旨在使学生了解数据驱动材料科学的内涵及趋势;理解和掌 握信息学的基本方法和代码;具备应用信息学手段解决材料设计问题能力,是材料专 业研究生获得材料信息学基础技能、接轨国际研究热点的有效途径。 *课程简介 (English) Course In recent years, big data and artificial intelligence, as a new
Description discipline, have been rapidly developed and popularized, which is embodied in Material Informatics in material science. It has become the trend of research mode development of material science and Engineering in the future, which is conducive to accelerating the research and development of materials and reducing costs. This course is offered to graduate students to prepare them for the paradigm shift. The main contents of the course include data-driven material science overview, database and data standards, highthroughput computing, high-throughput experimental methods and use cases. Machine learning algorithms and cases will be systematically taught. The purpose of this course is to enable students to understand the connotation and trend of data-driven material science, understand and master the basic methods and codes of Informatics, and have the ability to solve material design problems by means of informatics. It is an effective way for graduate students majoring in materials to acquire basic skills of material informatics and connect with international research hotspots. *教学大纲 (中文) Syllabus 教学内容 授课学 时 教学方 式 授课教 师 数据驱动的材料科学概述 4 授课 汪洪 数据库与数据标准 2 授课 汪洪 高通量实验设计的思路、类型、相关 技术、数据处理方法 4 授课 张澜庭 高通量虚拟实验教学 2 上机 张澜庭 材料跨尺度/多尺度计算概述 6 授课 张澜庭 几类基础性能的算法、高通量计算实 现方法 2 授课 张澜庭 高通量计算案例及上机实验 2 上机 张澜庭 期中考试 2 考试 汪洪 机器学习概述、模型评估 2 授课 鞠生宏 线性回归、支持向量机、聚类、降维 理论学习 4 授课 鞠生宏 线性回归、支持向量机、聚类、降维 上机练习 2 上机 鞠生宏
决策树、蒙特卡洛树理论学习 2 授课 鞠生宏 决策树、蒙特卡洛树上机练习 2 上机 鞠生宏 贝叶斯优化、神经网络理论学习 4 授课 鞠生宏 贝叶斯优化、神经网络上机练习 2 上机 鞠生宏 机器学习在材料设计中的应用 4 授课 鞠生宏 期末考试 2 考试 鞠生宏 *教学大纲 (English) Syllabus Content Hours Format Instructor Overview of data-driven materials science 4 Lecture Hong Wang Materials database and standards 2 Lecture Hong Wang High-throughput experiments 4 Lecture Lanting Zhang Vitural high-throughput experiments 2 Computer experiment Lanting Zhang Overview of multi-scale computational materials science 6 Lecture Lanting Zhang High-throughput algorithms 2 Lecture Lanting Zhang High-throughput computational experiment 2 Computer experiment Lanting Zhang Mid-term exam 2 Mid-term exam Hong Wang Overview of machine learning and model evaluation 2 Lecture Shenghong Ju Linear regression, support vector machine, clustering, dimensionality reduction 4 Lecture Shenghong Ju Computer experiment for linear regression, support vector machine, clustering, dimensionality reduction 2 Computer experiment Shenghong Ju
Decision tree, Monte Carlo tree search 2 Lecture Shenghong Ju Computer experiment for decision tree, Monte Carlo tree search 2 Computer experiment Shenghong Ju Bayesian optimization, neural network 4 Lecture Shenghong Ju Computer experiment for Bayesian optimization, neural network 2 Computer experiment Shenghong Ju Applications of machine learning in materials design 4 Lecture Shenghong Ju Final exam 2 Final exam Shenghong Ju *课程要求 (中文) Requirements 课程考核方式: 1). 平时作业,上机实验,占总分 30% 2). 期中考试,覆盖大约 50%次课内容,占总分 30% 3). 期末考试,覆盖大约 50%次课内容,占总分 40% *课程要求 (English) Requirements 1). Homework and computer experiment, 30% 2). Mid-term exam, 30% 3). Final exam, 40% 课程资源 (中文) Resources 课程资源 (English) Resources 备注 Note