Neuro-Symbolic AI
Fall 2025, CMU 10747

1. Website: http://www.cs.cmu.edu/~pradeepr/747
2. Piazza: https://piazza.com/cmu/fall2025/10747
3. Gradescope: https://www.gradescope.com/courses/1099796

Pradeep Ravikumar

Ayudh Saxena

Kaavya Subramanian

Course Description

The course explores what is arguably the next frontier/paradigm in AI: Neuro-Symbolic AI, which combines the modeling strengths of statistical/neural/deep learning with the reasoning and structured representational strengths of symbolic systems. Future breakthroughs in reliable reasoning will most likely rely on such an approach. An example of Neuro-Symbolic AI already in use is AlphaGeometry, an AI capable of solving Olympiad-level geometry problems.

The course will be structured as a partial seminar course. Of the two classes each week, one class will be a traditional lecture, while the other class will be a student-led reading-seminar on key papers, centering around the topic of that week's lecture.

Prerequisites

Basics of probability and statistical inference, and basics of machine learning (such as regression, classification, clustering). This course is for students who have already taken introductory courses in machine learning and statistics.

Course Goals

Class Website:

Discussion, Announcements:

Gradescope:

Textbooks

Schedule

Resources

Lecture Notes

1. Lecture 1: NSAI Intro & Logistics
2. Lecture 1: NSAI Overview
3. Lecture 2: NSAI Models

Course Policies

• Positive Advocate: The paper has not been published yet and is currently submitted to NeurIPS where you’ve been assigned as an emergency expert reviewer. Advocate for the paper to be accepted.
• Negative Advocate: The paper has not been published yet and is currently submitted to NeurIPS where you’ve been assigned as an emergency expert reviewer. Advocate for the paper to be rejected.
• Jury: You are to synthesize the pros and cons, decide whether the positive camp or the negative camp wins: whether the paper should be accepted or not, and why. They should also provide a written summary after class to the Reporter group to help with their report.
• Visitor from the Past: You are a visitor from [fill-in-the-blank] year in the past. How do you comprehend the results of the paper, in light of what you know? What surprises you most, and what do you most like or dislike? Would this have changed the course of ML history?
• Visitor from the Future: This paper was found buried under ground in the desert. You will determine where this paper sits in the context of previous and subsequent work. You must find and report on atleast one older paper cited within the current paper that substantially influenced the current paper and atleast one newer paper that cites this current paper. Keep an eye out for follow-up work that contradicts the takeaways in the current paper. This will comprise of two sub-groups, each focusing on different sets of papers.
• Academic researcher: You are a researcher who is working on a new project in this area. You are to propose potential follow-up projects not just based on the current paper but also only possible due to the existence and success of the current paper.
• Industry Practitioner: You work at a company or organization developing an application or product of your choice (that has not already been suggested in a prior session). Bring a convincing pitch for why you should be paid to implement the method in the paper, and discuss at least one positive and negative impact of this application.
• Private Investigator: Find out background information on one of the paper authors. Where have they worked? What did they study? What previous projects might have led to working on this one? What do you think motivated them to work on this project? Feel free to contact the authors, but remember to be courteous, polite, and on-topic. Write that you're in this seminar and include a link to this page.
• Reporter: You are to synthesize the discussions: the point not being to simply write down what everybody said or to summarize it all, but to provide higher level synthesis. This synthesis, in the form of a writeup, will be made available to all.

Order of presentations:

The positive and negative advocates (order between the two chosen at random in each class) -> Discussion between them and the Jury, also open to rest of the class -> Jury Decision -> Private Investigator -> Visitor from the Past -> Visitor from the Future -> Academic Researcher -> Industry Practitioner

Our hope is that this will allow you to do "deep" rather than shallow learning of the material (deep learning: not just for artificial neurons!)

Deliverables:

Before each paper discussion, each student should submit their pre-discussion sheet containing (via gradescope, due 11:59 PM the day before paper discussion):

• A new title for the paper and/or a new name for the algorithm it proposes.
• At least one question about the paper (either something you're confused about or something you'd like to hear discussed more)
• One paragraph on what they most liked about the paper.

Discussion Deliverables include:

• Each group (other than Jury, Reporter) will be presenting a few slides.
• The Jury group will be leading the in-class discussion comparing the positive and negative advocates. Within a day after class, they will also send a written summary to 10747-instructors@andrew.cmu.edu, which we will forward to the Reporter group
• The Reporter group synthesis report will be due in one week (same deadline as their next week pre-discussion sheet) via email to 10747-instructors@andrew.cmu.edu
• Each student will submit a very short intra-group peer evaluation form for each of their other group members for the previous week's seminar discussion, via a Google form that will be provided on Piazza, with the same deadline as their next week pre- discussion sheet

Extensions

Instead of homeworks and exams, we have a weekly cadence of paper discussions, which have their group based deliverables noted earlier. We also have a project with its timely deliverables. Due to the nature of these, extensions are not feasible. There are several exceptions:

• Medical Emergencies: If you are sick and unable to complete an assignment or attend class, please go to University Health Services. For minor illnesses, if necessary, we expect you to email us right away. For medical emergencies (e.g. prolonged hospitalization), students may request an extension afterwards and should include a note from University Health Services.
• Family/Personal Emergencies: If you have a family emergency (e.g. death in the family) or a personal emergency (e.g. mental health crisis), please contact your academic adviser or Counseling and Psychological Services (CaPS). In addition to offering support, they will reach out to the instructors for all your courses on your behalf to request an extension.
• University-Approved Absences: If you are attending an out-of-town university approved event (e.g. multi-day athletic/academic trip organized by the university), you may request an extension for the duration of the trip. You must provide confirmation of your attendance, usually from a faculty or staff organizer of the event.

For any of the above situations, you may request an extension by emailing 10747-instructors@andrew.cmu.edu – do not email the instructor or TAs. The email should be sent as soon as you are aware of the conflict and at least 5 days prior to the deadline. In the case of an emergency, no notice is needed.

Class Project

There will be a class project. You can form groups of up to two students. The projects all have a common meta-theme, which is neuro-symbolic AI for tabular data. Most representation learning work in ML focuses on images and text in part for historical reasons, and in part because these are the proto-typical "raw input" modalities, leaving tabular data to grapple with "classical" ML methods. It is up to you to correct this grave injustice! We will provide a candidate list of project topics (and more details) soon, but you are free to choose your own topic, in discussion with the instructional team.

1. Project proposal: Wednesday, October 8, 2025
2. Midway project report: Wednesday, November 5, 2025
3. Spotlight project presentations: In Class, December 1 and 3, 2025
4. Project reports in the style of NeurIPS paper: December 5, 2025

Grading

Audit Policy

Pass/Fail Policy

Accommodations for Students with Disabilities:

If you have a disability and are registered with the Office of Disability Resources, I encourage you to use their online system to notify me of your accommodations and discuss your 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.

Take care of yourself

• CaPS: 412-268-2922
• Re:solve Crisis Network: 888-796-8226

• On campus: CMU Police: 412-268-2323
• Off campus: 911