Course Syllabus
The syllabuses on both this page and the NTU online course information are synchronized.
Course Information
| Item | Content |
| Course title | Security and Privacy of Machine Learning |
| Semester | 114-1 |
| Designated for |
Intelligent Medicine Program GRADUATE INSTITUTE OF NETWORKING AND MULTIMEDIA GRADUATE INSTITUTE OF COMPUTER SCIENCE & INFORMATION ENGINEERING DEPARTMENT OF COMPUTER SCIENCE & INFOR |
| Instructor | SHANG-TSE CHEN |
| Curriculum No. | CSIE 5436 |
| Curriculum Id No. | 922 U4630 |
| Class | |
| Credit | 3 |
| Full/Half Yr. | Half |
| Required/Elective | Elective |
| Time | Wednesday 2,3,4(9:10~12:10) |
| Place | 資105 |
| Remarks |
Course Syllabus
| Item | Content |
| Course Description | Modern machine learning models has reached and even surpassed human performance in many areas. However, many of the successful cases only hold in clean and controlled settings, which can be far from real scenarios. This course will introduce you to potential vulnerabilities of ML models. We will design and implement various attacks during model training and testing phases, as well as methods to make ML models more robust. We will also cover other important aspects of ML, including privacy and fairness. |
| Course Objective | In this course, we will learn the security and privacy risks of AI in different tasks and settings, design strong defenses, and design more robust and safe models. Topics include adversarial attacks, poisoning attacks, jailbreak, model and data privacy, and data protection from misuse. |
| Course Requirement | |
| Expected weekly study hours before and/or after class | 3 |
| References | |
| Designated Reading |
Progress
| Week | Date | Topic |
Makeup Class Information
| NO | Date | Start Time | End Time | Location or Method |
Grading
| NO | Item | Pc | Explanations for the conditions |
Adjustment methods for students
| Adjustment method | |
| Teaching methods | |
| Assignment submission methods | |
| Exam methods | |
| Others |
Office Hour
| NO | Day | Start time | End time |
| Remarks | Wednesday 12:10-13:10 (or appointment via email) |