CS 7005: Advanced Special Topics in Computer Science

Instructor Information

Instructor: Chong Yu

Time: Tuesday & Thursday 2:00 - 3:20 PM

Location: Lindhall 0040

Office Hour: Tuesday 3:30 - 5:00 PM, Rhodes Hall 887

Course Description

This course addresses concerns about the reliability of machine learning technologies by exploring topics such as adversarial examples, model robustness, privacy, fairness, and federated learning. Students will develop skills to analyze vulnerabilities, enhance model resilience, and implement privacy-preserving techniques. The course also emphasizes understanding and mitigating biases to ensure fairness and explores the benefits of federated learning.

Workload

• 2 presentations over the semester
• 2 paper summaries over the semester

Grading

• In-class presentation: 50%
• In-class debate: 30%
• Discussion: 20%

Lateness Policy

• Paper presentations must be submitted by 6:00 PM the day before class. A 10% penalty applies for late submissions before the class begins. Submissions after the class receive a score of 0.
• Paper summaries must be submitted at the beginning of class. Late assignments are not accepted.

Electronic Submission: Please use Canvas to submit all materials.

Prerequisites

The class is primarily for graduate students in computer science. Senior undergraduates or students from other departments need professor's permission.

Schedule (Tentative)