Subject

Privacy, Security and Trust in Machine Learning Systems

1.Course titlePrivacy, Security and Trust in Machine Learning Systems
2.Codem23_w_055
3.Study programme—
4.Organizer of the study programme (unit, institute, department)Faculty of Computer Science and Engineering
5.Degree level (first, second, third cycle)Second cycle
6.Academic year / semester9 / Winter
7.Number of ECTS credits6
8.Teacher
9.Prerequisites for enrolling in the course—
10.Course objectives (competences)The goal of the course is introduction to the basic risks that may occur when applying machine learning in the software systems, the attacks that can be used to undermine their integrity, security and authority. the attacks that can be used to undermine their integrity, security and authority. In addition, strategies to protect systems from these most common attacks will be discussed. The second part of the course will address the challenges that exist with most systems that use machine learning, which is how to protect the privacy of the data used to train them. The final part of the course will focus on how to increase the confidence of machine learning systems by reviewing techniques for explaining their results.
11.Course content- Introduction to security dimensions, concepts, methods
- ML perspecrive on security
- ML solutions and their threats
- Attacks Against ML
- Choosing the right defense
- ML privacy issues
- Possible ML privacy solutions
- ML confidence issues and solutions
12.Learning methodsLectures supported by slide presentations, interactive lectures, practical classes (using equipment and software packages), teamwork, case studies, guest lecturers, independent preparation and defence of a project assignment and seminar paper, and learning in an electronic environment (forums and consultations).
13.Total available time6 ECTS x 30 hours = 180 hours
14.Distribution of the available time45 + 15 + 30 + 50 + 40 = 180 hours
15.Forms of teaching activities
15.1.Lectures – theoretical teaching45 hours
15.2.Exercises (laboratory, auditory), seminars, team work15 hours
16.Other forms of activities
16.1.Project assignments50 hours
16.2.Individual assignments30 hours
16.3.Home study40 hours
17.Grading
17.1.Tests45 points
17.2.Seminar paper / project (presentation: written and oral)50 points
17.3.Activity and participation10 points
17.4.Final exam0 points
18.Grading criteria (points / grade)
up to 50 points5 (five) (F)
from 51 to 60 points6 (six) (E)
from 61 to 70 points7 (seven) (D)
from 71 to 80 points8 (eight) (C)
from 81 to 90 points9 (nine) (B)
from 91 to 100 points10 (ten) (A)
19.Requirements for signature and taking the final examcompleted activities
20.Language of instructionMacedonian and English
21.Method of monitoring teaching qualityinternal evaluation and survey mechanism
22.Literature
22.1.Required literature
1. J. Morris Chang, Di Zhuang, G. Dumindu Samaraweera | Privacy-Preserving Machine Learning | Manning | 2023
2. Jin Li, Ping Li, Zheli Liu, Xiaofeng Chen, Tong Li | Privacy-Preserving Machine Learning | Springer | 2022
3. Yevgeniy Vorobeychik, Murat Kantarciogly | Adversarial Machine Learning | Springer | 2022
4. Anthony D. Joseph, Blaine Nelson, Benjamin I. P. Rubinstein, J. D. Tygar | Adversarial Machine Learning | Cambridge University Press | 2019
5. Tianqing Zhu, Gang Li, Wanlei Zhou, Philip S. Yu | Differential Privacy and Applications | Springer | 2017
6. Christoph Molnar | Interpretable Machine Learning | Leanpub | 2020
22.2.Additional literature
No. Author Title Publisher Year