Course Syllabus
Draft of April 22; subject to updates
Time and place
MW 5-6:20pm, CIC 1203
Course description
Cybersecurity has become a high-stakes battlefield, and machine learning is both a weapon and a shield. From home IoT devices to critical infrastructure, the stakes have never been higher: the attack surface of our systems keeps increasing; attackers are more and more determined and sophisticated; and ML is emerging as a tool to automate attacks. This course will explore the most significant uses of ML for cybersecurity over the past four decades while focusing on the most promising defenses proposed in recent years. Example applications that will be covered are malware detection, spam detection, anomaly detection for computer networks, anomaly detection for industrial control systems, generating private synthetic data, vulnerability detection, and assisting in security-relevant decision-making by both security experts and end users. The course will also teach students to think critically about the challenges and pitfalls of applying ML to cybersecurity, such as the base rate fallacy, unavailability of training data, lack of interpretability of ML models, and susceptibility of ML models to adversarial examples.
A preliminary list of papers (which will certainly be updated) that the course will cover can be found here. The list comprises seminal papers, recent award-winning papers, and cutting-edge new results.
Prerequisites
The class is aimed at PhD students, masters students, and undergraduates with a strong background in computer security, programming, and ML. Students must have taken (15/18-330 or 18-631 or equivalent systems-oriented security course) AND (10-601 or 18-661 or equivalent). Coursework will require downloading and running code from github, ML programming, reviewing and presenting research papers, and working on a semester-long course project.
Learning goals
Students will be able to:
- Describe the historical and proposed uses of ML for cybersecurity
- Analyze the practicality of proposed ML-based tools for cybersecurity
- Define and describe experiments to demonstrate that an ML-based tool adequately solves a cybersecurity problem
- Identify and describe the challenges of using ML to solve specific cybersecurity problems
Learning resources
The course has no textbook, only assigned readings. All the readings are available online, both from the publishers' web sites (which may require you to be on campus or using the CMU VPN) and on the course HotCRP instance (TBA).
Assessments
15% presentation: individually or in groups, students will present a paper and lead the class in a critical discussion
10% participation: students will participate in class discussions
25% paper critiques: students will submit brief critiques of assigned readings and occasionally be asked to discuss them in class
50% project: working individually or in groups, students will evaluate, extend, or develop a tool that uses ML to solve a cybersecurity problem
Attendance
Since this is a discussion-based class, in-person attendance is expected and is reflected in the participation grade. Up to three absences (when you are not the discussion lead) are allowed without penalty. If you will be absent or substantially late, please contact the instructor ahead of time.
Late and make-up work
Since the course does not have traditional homework assignments and deliverable submission dates for the entire semester will be publicized during the second week of classes, late work will be accepted only under special circumstances and with the instructor's approval.
Academic integrity
Paper critiques should be written by the students to whom they are assigned. The initial submitted version of a paper critique should not be discussed without others before submission; after submission it may be discussed with others and revised on the basis of that discussion. Other deliverables should be completed by the groups of students to whom they are assigned, but may be discussed with others at any point.
AI policy
The use of AI tools for assignments with written deliverables is allowed, with the following conditions:
- You must disclose every instance of using an AI tool, including prompt, if applicable.
- You must describe how AI generated content is related to the content you turned in or presented. What was completely accurate and what needed changing?
- You are responsible for the content (e.g., in discussion).
If you are not careful, using AI tools will cause you to learn less.
Health and wellness
Do your best to maintain a healthy lifestyle this semester by eating well, exercising, avoiding drugs and alcohol, getting enough sleep and taking some time to relax. This will help you achieve your goals and cope with stress.
All of us benefit from support during times of struggle. You are not alone. There are many helpful resources available on campus and an important part of the college experience is learning how to ask for help. Asking for support sooner rather than later is often helpful.
If you or anyone you know experiences any academic stress, difficult life events, or feelings like anxiety or depression, we strongly encourage you to seek support, including through CMU resources. Consider reaching out to a friend, faculty or family member you trust for help getting connected to the support that can help.
Accommodations for students with disabilities
If you have a disability and have an accommodations letter from the Disability Resources office, I encourage you to discuss your accommodations and needs with me as early in the semester as possible. I will work with you to ensure that accommodations are provided as appropriate. If you suspect that you may have a disability and would benefit from accommodations but are not yet registered with the Office of Disability Resources, I encourage you to contact them at access@andrew.cmu.edu.
We must treat every individual with respect.
We are diverse in many ways, and this diversity is fundamental to building and maintaining an equitable and inclusive campus community. Diversity can refer to multiple ways that we identify ourselves, including but not limited to race, color, national origin, language, sex, disability, age, sexual orientation, gender identity, religion, creed, ancestry, belief, veteran status, or genetic information. Each of these diverse identities, along with many others not mentioned here, shape the perspectives our students, faculty, and staff bring to our campus. We, at CMU, will work to promote diversity, equity and inclusion not only because diversity fuels excellence and innovation, but because we want to pursue justice.
Each of us is responsible for creating a safer, more inclusive environment.
Unfortunately, incidents of bias or discrimination do occur, whether intentional or unintentional. They contribute to creating an unwelcoming environment for individuals and groups at the university. The university and your instructor encourage anyone who experiences or observes unfair or hostile treatment on the basis of identity to speak out for justice and support, within the moment of the incident or after the incident has passed, including using the following resources:
All reports will be documented and deliberated.
Course Summary:
| Date | Details | Due |
|---|---|---|