To Students: This website is under construction; please check back frequently. Course logistics may change.

Overview: This course offers a comprehensive introduction to the principles of machine learning (ML) and deep learning (DL), emphasizing both mathematical foundations and practical applications. You will gain insights into basic ML techniques, learn advanced DL applications in fields such as computer vision and natural language processing, and understand their impact on areas such as image recognition and autonomous systems. The course includes a hands-on assignment and a customizable final project, giving you practical experience in implementing ML and DL solutions.

Prerequisites: Proficiency in Python; college calculus and linear algebra; probability and statistics. CSCI 567 (Machine Learning) or equivalent is recommended but is not a formal prerequisite. Duplicate credit is not given for CSCI 527.

Basic logistics:

• Course: CSCI 566 — Deep Learning and its Applications
• Units: 4.0
• Time: Monday, 5:00–8:20 PM
• Sections: 30122 (in person, SGM 124) and 30359 (DEN@Viterbi, remote). Both sections meet at the same time and receive the same lecture, with one exception: the August 31 meeting is a recorded lecture for both sections, with no live session and no meeting in SGM 124 that day. Where a policy differs between sections, the syllabus says so explicitly.
• Instructor: Yue Zhao
• Office Hours: During and after the Monday lecture block. DEN students may join remotely.
• Teaching Assistants: Tiankai Yang, Jiate Li, Chen Chu. TA office hours: TBD.
• Discussion: Piazza
• Submissions: Gradescope. The entry code is announced on Piazza and in the first lecture.
• Slides: Google Drive, posted as PDF after each lecture. Sign in with your USC account — the folder is shared with USC only, so a personal Gmail account will see a request-access screen instead of the slides.
• Syllabus: syllabus.pdf

Assessment: Two in-class quizzes 10 points (5 each), one assignment 10 points, midterm 35 points, and course project 45 points, for 100 points total. There is no final exam. Quiz dates are announced at least one week in advance. Within the project, the poster defense and an anonymous peer evaluation are graded individually, so members of the same group can receive different project grades. See the syllabus for the full breakdown, including the separate path for DEN students, who do not attend the poster session.

Course Format: Fourteen Monday meetings. Lectures combine mathematical foundations with implementation, and the semester project runs from a one-page pre-proposal through a midterm report to a final report and a poster session in the last meeting. Materials and submissions are managed through Brightspace and Gradescope.

Guest Lectures: Industry and academic professionals join our lectures regularly, sharing their experience in ML and data science and providing career insights.