课程号：

TE196002

课程名：

强化学习导论

英文课程名：

Introduction to Reinforcement Learning

开课单位：

通信工程学院

授课教师：

李葆春

责任教师：

顾华玺

助理教师：

余晓杉

学分/学时：

1/16

选课年级：

大二或大三学生

选课专业：

计算机、通信专业

教师简介（中英文）：

李葆春教授为计算机网络界著名学者，IEEE fellow，已发表学术论文300余篇，被引用次数高达16762次，H指数为75，i10指数229。2000年获得IEEE通信系统领域Leonard G.Abraham奖，2009年分别获得IEEE通信协会多媒体通信最佳论文奖和多伦多大学McLean奖。李葆春教授在大规模网络系统、数据中心网络、网络编码、网络通信机制等领域开展系统且深入的研究并做出了突出贡献，推动了学术研究和产业发展，其成果被广泛引用并赢得了国际赞誉。

Baochun Li is a well-known scholar in the area of computer networks. He has been elevated to the grade of IEEE Fellow, and he has co-authored more than 320 research papers, with a total of over 15000 citations, an H-index of 73 and an i10-index of 220. In 2000, he was the recipient of the IEEE Communications Society Leonard G. Abraham Prize Paper Award in the Field of Communications Systems, In 2009, he won the Multimedia Communications Best Paper Award from the IEEE Communications Society and the University of Toronto McLean Award. Baochun’s research interests are in the area of large-scale distributed systems, cloud computing, datacenter networks, applications of network coding, mobile computing, and wireless networks. His research results have been cited widely and won international acclaim.

课程简介（中英文）：

本课程的主题是强化学习，它是机器学习的一个子领域。强化学习旨在解决复杂环境下最佳行为学习过程中遇到的普遍性和具有挑战性的问题。该学习过程仅由奖励值以及从环境中获得的观察结果驱动。该模型具有很强的通用性，可以应用于许多实际情况，从玩游戏（著名的AlphaGo示例）到云资源分配优化。由于其灵活性和通用性，强化学习领域发展非常迅速，吸引了大量学术界和商业界研究人员的关注。本课题旨在弥补强化学习技术实践与理论之间的差异。一方面，近来强化学习的研究结果过于专业化和抽象化，读者需要通过大量的学习才能理解；另一方面，读者很难将书籍或研究论文中侧重于数学形式描述的理论方法转化成能够解决实际问题的工作实现。这就使得对强化学习领域感兴趣的人难以直观地了解强化学习中的方法和思想。本课程侧重于实践而非理论知识，通过引入清晰易懂的具体编程实例来解决以上问题。编程实例由当今最流行的机器学习框架之一，PyTorch框架来实现，该框架具有强大的功能和表现性。

The topic of this course is Reinforcement Learning, which is a subfield of Machine Learning. Reinforcement learning is designed to solve the general and challenging problem of learning optimal behavior in a complex environment. The learning process is driven only by reward values and observations obtained from the environment. This model is very general and can be applied to many practical situations from playing Go games (in the famous example of AlphaGo) to optimizing resource allocation in the cloud. Due to its flexibility and generality, the field of Reinforcement Learning is developing very quickly and attracts a substantial amount of attention both from both academic and industry researchers. This course is designed to fill the gap between practice and theory on Reinforcement Learning techniques. On one hand, recent research results on Reinforcement Learning is too specialized and abstract to be understandable without serious efforts. On the other hand, it is not always obvious how to make a step from the theoretical methods described in mathematical-heavy forms in a book or a research paper to a working implementation solving actual problems. This makes it hard for somebody interested in the field to get an intuitive understanding of methods and ideas in Reinforcement Learning. With a focus on practice rather than theory, this course rectifies this problem by introducing concrete programming examples that are clean and easy to understand, made possible by the expressiveness and power of the PyTorch framework, one of the most popular modern frameworks for machine learning today.

先修课程：

无