Tsinghua Computer science as at Jan 2019 -faculty
introduces the advanced theory of machine learning and its related algorithms. The course will first review the state-of-the-art
machine learning algorithms and the course’s content mainly consists of probabilistic generative learning and probabilistic
discriminative learning. Based on the theoretical analysis and algorithmic application, we plan to introduce the following
（1） Probabilistic topic model
（2） Restricted Boltzmann machines
Factor graph model
（4） Bayesian nonparametrics
（5） Semi-supervised learning
Scalable machine learning
The course requires all students to design and implement an algorithm for advanced machine
learning, and validate the algorithm on the our provided platform.
|Jie TANG, Jun ZHU||Data representation(e.g.,vector space model,language model); Basic probability theory(e.g., likelihood,conditional probability,posterior
probability,Bayes);Basic linear algerbra(e.g.,linear transformations,eigenvalues,least-squares best fit)|
This is an introductory course
on distributed systems. This course introduces the principles of distributed systems as well as some of the current influential
large-scale distributed systems such as Google file system, MapReduce, Amazon Dynamo etc. To make the course more concrete,
this course uses a series of labs requiring the students to build real distributed systems. This course emphasizes on the
general principles of building distributed systems in addition to introducing important practical distributed systems. For
example, the various kinds of distributed consistency protocols will be discussed and such principles can be adopted in many
kinds of real distributed applications. The current systems used by Googe, Amazon, Microsoft will be introduced.
|Kang CHEN||data structure, C programming||Class attendance is required, Readings and Homework (20%), Labs (40%), Final Exam (40%). Exam will be
This course discusses the basic design principles of the Internet and where the future Internet architecture
is heading to. It is for graduate students and we will cultivate the research capabilities of students by this course. 10
lectures will be given. Students are required to read papers and give presentations in the class, as well as finish the course
project. The course language is English.
|Dan LI||-||Class attendance is required, Paper reading presentations
This course covers the basic understanding
of human perception and cognition, interaction styles development, design and evaluation of GUI, and natural human computer
interface technologies. Input technologies are emphasized. Multimodality about visual, acoustic and touch sense channels are
introduced with new input interfaces. Signal processing, feature extraction, and mapping schemes will also be covered. Measure
methods are for the efficiency of interaction. Hands-on laboratories, study reports, individual projects and semester long
group projects will be assigned, some can potentially continue as further researches.
|Yuanchun SHI||Programming||Class attendance is required, Assignments|
Software project management is the art and science
of planning and leading software projects. It is a sub-discipline of project management in which software projects are planned,
monitored and controlled. Using software development life cycle and process groups as two main threads, this course introduces
the procedures and methods of software project management, which will help keep projects under control, and produce software
with required functions in shorter time, higher quality and predicable cost. This course follows the nine Project Management
Knowledge Areas - project integration, scope, time, cost, quality, human resources, communications, risk, and procurement
- using the experiences of real-life businesses. The students are organized in groups to do the homework and presentations.
Virtual software projects are assigned to groups. Typical scenarios are designed in the projects to help students understand
the nine knowledge areas by solving the issues in the projects through group discussion and collaboration.
|Yong ZHANG||Programming skill||Class attendance is required, Assignments (20%), Mid-term report (15%), Final short show (35%), Final open-book examination
The course starts with an overview of the fundamental concepts, algorithms
and tools in multimedia processing, including information theory, signal processing, signal compression, and communications.
Then, a brief introduction of multimedia related standards such as H.264, HEVC, RTP are provided. Combining the fundamentals
and the standards, we will finally introduce the students to challenges and current solutions to various challenges in large
scale, real time and mobile multimedia applications.
|Jiangtao (Gene) WEN||-||Attendance/quizzes (25%), Projects/presentations
(15%/5% + 15%/5% + 25%/10%)|
This course gives a survey to the new research branches, introduces the state-of-the-art technologies,
and discusses on open problems and challenges on Web information retrieval (Web IR). At the same time, the course focuses
on the real applications in the Internet environment, making case study and detail analysis on commercial search engines (SE).
The main topics of the course includes (but not limited to): IR in Web environment, such as link analysis, anti-spam, etc;
question answering; opinion / sentimental analysis; social media and IR; personalized IR and recommendation; user behavior
analysis; online advertisement; mobile search; and IR and SE evaluations. The course is composed of lectures and student-conducted
|Min ZHANG||Programming skill||Class attendance is required, Assignments, Projects and Tea Time Presentations (bonus score)|
The course starts with an
overview of the big data analytics, clustering and distributed programming. We will also cover methods for processing big
data as well as optimization techniques. Graph processing and visualization of big data will be covered. There will be labs
and projects which allow students to experiment with real data and apply the knowledge of what they learnt in class.
|Wenguang CHEN||Some programming knowledge||Assignments and Labs (50%), Projects (15%), Mid-Term Exam (10%), Final Exam (25%).|
This course is a graduate
course and is primarily project-oriented. It will cover three major of aspects of IP network management: networks, objectives
and methodologies. There will be 12 lectures given. Students are expected to form a team of two and finish a project on the
THU-INM (Tsinghua University IP Network Management) platform.
|Dan PEI||C Programming, Undergraduate Network
of Projects (50%), Technical Reports (30%), Presentation (20%).|
The course introduces natural
language processing (NLP), from its history to recent advances in deep learning applied to NLP. NLP is one of the most important
technologies in Artificial Intelligence. NLP aims at enabling computers to understand human languages and communicate with
humans. There are a large variety of tasks and machine learning methods in NLP. In this course, we plan to introduce the following
（1） The history and the tasks in NLP.
（2） Basic tasks in NLP: Sequence
tagging, parsing, classification and clustering.
（3） Applications in NLP: machine translation, question
（4） Recent advances in deep learning applied to NLP.
（5） Open problems
and challenges for NLP.
By learning from lectures and programming assignments, students will master necessary knowledge
about NLP and engineering tricks for practical NLP problems.
|Zhiyuan Liu||basic probability
theory; basic linear algebra; programming skills (C++ or Python). Knowledge of machine learning is preferred.||Class attendance is required, assignments,
Theoretical Computer Science
course introduces the basic concepts and history of deep learning as well as recent developments. The course will cover the
theoretical foundation and typical models of deep learning and discuss the cutting-edge research about deep learning. Specifically,
we will introduce the following topics:
（1） Basics of machine learning
（3） Convolutional neural networks
（4） Long short-term memory networks
（5） Deep generative models
（6） Generative adversarial networks
libraries (e.g., Tensorflow, ZhuSuan)
The course requires all students to design a deep learning model for solving a
practical problem, which can be chosen from a set of problems provided by the teacher or proposed by yourself.
level of probability theory, linear algebra and calculus. Basic level of programming skills.||Class attendance is required, Assignments, Exams/Projects.|xxHot
Topics in Computational Biology
The course will cover the following topics: NP completeness, PSPACE, L Space, IP system,
BPP, derandomization, PCP, classical communication complexity, circuit complexity, Decision tree complexity. quickly recall
basics about convex optimization and machine learning: linear/logistic regression, regularization, newton method, stochastic
gradient descent (asychronized, variance reduction method), generative vs discrimitive, variance vs bias. Off-the-shelf machine
learning and prediction algorithms: k-nn, SVM, kernel trick, clustering, Adaboost, gradient boosting, random forest. Online
learning and sequential prediction. Multi-armed bandit, Universal portfolio, Multiplicative weighting method, online convex
optimization, basic time series. linear algebra-based learning algorithms: SVD, principle component analysis (PCA), independent
component analysis (ICA), Nonnegative matrix factorization (NMF), topic modeling, matrix completion, dictionary learning,
tensor method, spectral clustering.
|Andrew YAO||Discrete Mathmatics||Assignments and Final Exam|
The course covers research progress and hot topics in Computational Biology and introduces
topics including basic computational theory and methods, three-dimensional structure determination and dynamic study of proteins,
protein and drug molecular design, Proteomics, and Biology evolution model.
|Jianyang ZENG||Machine Learning||Projects and Final Exam|| || || || || |