Spring 2022

Semester Starts!

The spring quarter starts on March 28, 2022.

ECS 289G Advanced Deep Learning (syllabus)

Course Information

ECS 289G Advanced Deep Learning
Date: March 28-June 2, 2022
Time: M/W 3:10-4:30PM
Virtual Venue: Zoom Meeting

Instructor Information

Instructor: Jiawei Zhang
Email: jiwzhang@ucdavis.edu
Office: Zoom Link to Join
Office Hours: M 4:30-5:30PM

Teaching Assistant: Xiao Liu
Email: xioliu@ucdavis.edu
Office Hours: R 2:00-3:00PM
Office: Zoom Link to Join

Course Description

The course on Advanced Deep Learning focuses on selective areas of importance about deep learning. Deep learning has been one of the hottest topics in AI studies, and techniques developed in the research hold substantial impacts in many important applications. Selective topics will be covered in the Advanced Deep Learning.

Topics Covered

  • Deep Learning
  • Computer Vision
  • NLP
  • Graph Mining
  • Recommender System
  • Robotics


No required textbook.


You are expected to have background knowledge in Data Structure, Algorithm, Discrete Mathematics. You will also need to be familiar with basic Linear Algebra, basic Statistics, and can master at least one programming language and have programming experiences.

Course Format

The objective of this course is to familiarize students with the latest research topics related to deep learning. Course activities include 1) paper reading and paper review; 2) paper presentation and discussion; and 3) research oriented course paper writing.

  • Paper Reading and Paper Review : Each class will discuss one academic paper. Before class, the students should read the paper to be presented in class, and write a short review (no longer than 1 page) for the paper. The review should cover: (a) a summary of the paper; (b) 3 strong points of the paper; (c) 3 weakness of the paper; (d) potential ideas of future works; and (e) questions about the paper you would like to ask the presenter. Students need to submit the review before class starts to the Canvas system. Some external reference papers/articles can be provided for students to understand the papers presented in class (no need to present or write reviews for external reference papers).
  • Paper Presentation and Discussion : Students may form a group (max size 4) to present a paper, and students will not need to write the reviews of papers you choose to present. Paper presentation is expected to take about 1 hour to cover (a) background knowledge, (b) problem description, (c) definitions, (d) ideas, (e) proposed methods and techniques, (f) experiments, (g) results, (h) related works, and (i) potential future works, etc. Too short presentations will be penalized. Additionally, we will have 15-20 minutes for in-depth discussion about the paper and Q&A with the audiences. Students can have Q&A during and after the presentation. Students need to send the slides before the end of presentation day to the TA.
  • Paper Proposal and Paper Writing : Each group needs to finish an independent research-oriented academic paper in this course, based on the problems studied in this course about deep learning. The paper should be original work, not recycled published/submitted/on-going work with another faculty or classes. Students need to submit a paper proposal (no longer than 2 pages, ACM sample-sigconf double-column conference format) at the mid of the class to provide the planning for the paper. By the end of this quarter of this course, students need to submit the final paper (10 pages in ACM sample-sigconf double-column format). Students need to submit the proposal and final paper before the deadlines. The submitted paper will be graded according to the novelty, technical depth, completeness and research value (low-quality papers will be penalized as well).

Course Resources

Presentation Schedule and Progress

Schedule Sheet Link

Reference List


10/10 Weeks


19/19 Papers

Final Paper

Final Paper Due on June 3

Grading Policy

In-class presentation: 30% . Powerpoint presentation needs to be submitted on the day of the presentation, before 11:59PM (midnight) of your presentation day. Copying existing presentation from the web is regarded as plagiarism.
Course participation and QA: 20% . A summary/review of each in-class discussion paper needs to be submitted before each class starts. During the class, presenter and audiences can have Q&A with the pre-prepared questions in the review report.
Course paper: 50% . An original work on deep learning. Not recycled published/submitted/on-going work with another faculties or classes.

  • Paper proposal: 10%.
  • Final paper: 40%.
Final Grade
  • A+ [100, 97], A: (97-93] A-: (93-90];
  • B+ (90-87], B: (87-83], B-: (83-80];
  • C: (80-70];
  • D: (70-60];
  • F: (60-0];

Late Submission Policy

  • Late paper review submission will not be accepted.
  • Late presentation slides submission will not be accepted.
  • Late proposal submission will not be accepted.
  • Late final paper submission will get your grade * 1(t<=24)e^(-(ln (2)/12) t) , if you are t hours late.

Other Issues

Gemeral Policy

  • University Attendance Policy: Excused absences include documented illness, deaths in the family and other documented crises, call to active military duty or jury duty, religious holy days, and official University activities. These absences will be accommodated in a way that does not arbitrarily penalize students who have a valid excuse. Consideration will also be given to students whose dependent children experience serious illness.
  • Academic Honor Policy: The UC Davis Code of Academic Conduct outlines the University's expectations for the integrity of students' academic work, the procedures for resolving alleged violations of those expectations, and the rights and responsibilities of students and faculty members throughout the process. Students are responsible for reading the Code of Academic Conduct and for living up to their pledge to "...be honest and truthful and... [to] strive for personal and institutional integrity at UC Davis." (UC Davis Code of Academic Conduct, found here)
  • Syllabus Change Policy: Except for changes that substantially affect implementation of the evaluation (grading) statement, this syllabus is a guide for the course and is subject to change with advance notice.
Collaboration/Academic Honesty: All course participants must adhere to the academic honor code of UC Davis which is available in the student handbook. All instances of academic dishonesty will be reported to the university. Every student must write his/her own homework/code (unless you are in the same group for the programming project). Showing your code or homework solutions to others is a violation of academic honesty. It is your responsibility to ensure that others cannot access your code or homework solutions. Consulting related textbooks, papers and information available on the Internet for your coding assignment and homework is fine. However, copying a large portion of such information will be considered as academic dishonesty. If you borrow a small piece of any such information, please acknowledge that in your assignment. Please see the following link for a complete explanation of the Code of Academic Conduct.
Student Disabilities: Americans With Disabilities Act: Students with disabilities needing academic accommodation should: (1) register with and provide documentation to the Student Disability Center; (2) bring a letter to the instructor indicating the need for accommodation and what type. This syllabus and other class materials are available in alternative format upon request. For more information about services available to UC Davis students with disabilities, contact the Student Disability Center.