Machine Learning in Cybersecurity Mario Fritz

News

14.07.2020

October Exam date Preferences

Dear Students,

We hope you are all well during these extraordinary times.

As announced earlier the remaining exams are postponed to the examination
slot right before the winter term. These are the two dates that we
going to offer:

29 & 30 October... Read more

Dear Students,

We hope you are all well during these extraordinary times.

As announced earlier the remaining exams are postponed to the examination
slot right before the winter term. These are the two dates that we
going to offer:

29 & 30 October 2020

Please use the form below to enter your exam date preference:

https://forms.gle/vFJ1tTgJKkuTnuQf8

If you are not comfortable with the google form, you can alternatively email your information and preferred date(s) to:

mlcysec_ws1920_staff@lists.cispa.saarland

with subject:

"MLCySec1920 Oct. Exam Date Preference - FirstName LastName"

30.01.2020

Final Project: Report and Presentations

Report

  • The final project report is due on 4-Feb 23:59:59. Please email us the report (pdf).
  • Please follow the double column template mentioned in the handout. The report (excluding references and appendix) should not exceed 4 pages.

Project... Read more

Report

  • The final project report is due on 4-Feb 23:59:59. Please email us the report (pdf).
  • Please follow the double column template mentioned in the handout. The report (excluding references and appendix) should not exceed 4 pages.

Project Presentations

  • The final project presentations take place during next week's regular lecture+tutorial slot: 6-Feb 14:15 onwards.
  • Please email us your presentations (preferably pdf; otherwise ppt, key, google drive url) before 6-Feb 14:00. We will project it using our laptop.
  • The duration of the presentation should be at most 10 minutes. This will be followed by a short Q&A.

Additional details available in today's lecture slides.

28.01.2020

Register for Exam

By now, you should have received your examination slots by email (the address entered in the google form).

Important:

  • Let us know immediately if you spot any errors, or haven't received the email!
  • Please separately register with the examination office by... Read more

By now, you should have received your examination slots by email (the address entered in the google form).

Important:

  • Let us know immediately if you spot any errors, or haven't received the email!
  • Please separately register with the examination office by 31-Jan-20.
     
17.01.2020

Advertisement: CISPA Young Researcher Security Convention SeCon 2020

Application is open and seats are still available: https://cispa.saarland/secon/

 

Machine Learning in Cybersecurity

 

Course Description

Recent advances in Machine Learning has lead to near (or beyond) human-level performance in many tasks - autonomous driving, voice assistance, playing a variety of games. In terms of privacy and security, this is a double-edged sword. ML techniques can be used to efficiently detect and prevent attacks (e.g., intrusion detection). However, their deployment to many real-world sensitive systems (e.g., self-driving cars, the cloud) also makes them susceptible to numerous attacks, such as introducing imperceptible perturbations in inputs and forcing ML systems behave in unintended ways.

The course explores in-depth both of these sides to Machine Learning and Cyber Security. The content addresses the following areas:

  1. ML overview
  2. ML for improving security
  3. Attacks on ML models
  4. ML and Privacy

 

Class Time and Location

Winter Semester 2019/20
Lectures: Thursday, 14:00 - 16:00  (starts at 14:15)
Exercises: Thursday, 16:00 - 18:00  (starts at 16:15)
Location: E91, 0.05

Instructor: Mario Fritz
Teaching Assistants: Tribhuvanesh Orekondy, Hossein Hajipour, Shadi Rahimian, Dingfan Chen

Contact us at: email

 

Schedule and Syllabus

Event Type Date Description Course Materials
Lecture October 17 Logistics and Course Overview Slides
Tutorial October 17 Primer on Python and Jupyter Slides | Notebook
Lecture October 24 ML Overview Slides
Tutorial October 24 Scikit-learn, and Pytorch basics Notebook
Lecture October 31 DL overview Slides
Tutorial October 31 Applying Deep Learning Notebook
Lecture Nov 7 Anomaly Detection Slides
Tutorial Nov 7 (No tutorial)  
Project 1   (Due 13-Nov 23:59) Notebook
Lecture Nov 14 Evasion Attacks Slides
Tutorial Nov 14 GPU cluster, Project 2 Handout  
Lecture Nov 21 Poisoning Slides
Tutorial Nov 21 Project 1: Summary and Discussion Slides
    Using the GPU Machines Slides
Project 2   (Due 28-Nov 13:59)

Notebook

Lecture Nov 28 Game Theory, Membership Inference Slides
Tutorial Nov 28 Project 3 Handout  
Project 3   (Due 12-Dec 13:59) Notebook
Lecture Dec 5 (No Lecture)  
Tutorial Dec 5 Project 2: Summary and Discussion Slides
Lecture Dec 12 Model Stealing, Water Marking Slides
Tutorial Dec 12 (No Tutorial)  
Project 4   (Multiple deadlines - see handout) Handout (v1.0)
Lecture Dec 19 GAN, Malware (partial)

Slides_GAN, Slides_Malware

Lecture Jan 9 Malware (cont), Differential Privacy Slides_Malware, Slides_DP
Lecture Jan 16 Differential Privacy (continued) Slides
Lecture Jan 23 (No Lecture)  
Lecture Jan 30 Federated Learning Slides
Lecture Feb 6 Final Project Presentations (replaces lecture+tutorial)  

 

Course Discussions

We encourage you to subscribe to the course mailing list via this interface.

mlcysec_ws1920_stud@lists.cispa.saarland

Using this mailing list, you can reach out to other students for course-related discussions or finding team-mates.

 

Prerequisites

  • Programming: Basic programming skills and familiarity with Python. All assignments will be in Python.
  • Machine Learning: Prior knowledge in ML is helpful.
  • Linear Algebra, Probability, Statistics and Calculus: Introductory level.

 

Project Honor Code

We encourage discussing ideas and concepts with other students to help you learn and better understand the course content. However, the work you submit and present must be original and demonstrate your effort in solving the presented problems. We will not tolerate blatantly using existing solutions (such as from the internet), improper collaboration (e.g., sharing code or experimental data between groups) and plagiarism. If the honor code is not met, no points will be awarded.



Privacy Policy | Legal Notice
If you encounter technical problems, please contact the administrators