Course syllabus

Jump to today

Course-PM

DAT615 Neuro-symbolic AI lp2 HT26 (7.5 hp)

Course is offered by the department of Computer Science And Engineering

Contact details

For questions about their lectures, contact the respective lecturer. For questions about assignments, contact the TA. Note that email is the preferred mode of contact, not Canvas DMs. The teachers will monitor their emails much more regularly than Canvas DMs.  

Student Representatives

TBA

Course purpose

The purpose of the course is to introduce the students to the field of neuro-symbolic AI and bring them to a level where they can gain an understanding of the research frontier. Therefore, the course will slightly shift each year to accommodate the rapidly developing landscape of AI. Our aim is to give the students the tools and abilities to continue following the research frontier, even after the course.

Recent advances in neural networks have quickly moved the state of the art forward for many applications, not least natural language processing and computer vision. But neural models still have weaknesses, for instance when it comes to reliably reasoning logically or planning. While Large Language Models are improving, reasoning is were methods from symbolic AI (e.g. rule based, logical methods) still provide the guarantees and reliability that neural methods lack. The aim of neuro-symbolic AI is to combine the two approaches to get the best from both worlds. It will also allow us to tackle problems which neither method on its own is sufficient. In this course, we will focus on two application areas: cognitive modeling and languages for emergent communication (with Dubhashi), and mathematics and programming (with Johansson). Towards the end of the course, students should have sufficient knowledge to be able to follow recent research papers in these areas. The final two weeks are dedicated to presentation and discussion of such state of the art research articles. 

Students should expect an advanced course keeping in line with state of the art research in neuro-symbolic AI. Students should expect to do a fair bit of reading of advanced scientific literature. This should be understood to be an advanced MSc-level course, requiring corresponding effort.

To help students select courses the following might help to align your expectations with what will be the contents of this course.

DO take this course if:

  • You are interested in the latest research in neuro-symbolic AI, and prepared to accept that the course may therefore vary slightly from year to year.
  • You enjoy attending lectures delivered by active researchers in the field (all teachers in this course do research on neuro-symbolic AI).
  • You are prepared to do a lot of reading of scientific literature, and learn how to digest and present modern research papers.
  • You are willing to ask technical questions and willing to work to understand advanced scientific concepts.

DO NOT take this course if:

  • You want a course based on a single textbook, with little variation from year to year.
  • You want a course primarily based on programming exercises.
  • You do not want to attend lectures. Historically, students who attend the lectures have performed considerably better in the final exam.
  • You want to do as little work as possible to pass/get a certain grade. Some assignments in this course are voluntary, so self-motivation and a genuine interest in the topic is essential.

Schedule

Please refer to TimeEdit for details of the schedule. Note that rooms might vary between lectures (for reasons beyond our control). 

Below is an overview of the lecture topics. More details will be added shortly.

Week   Tuesday 10:00    Friday 10:00 
1 3 Nov

Introduction to neuro-symbolic AI 

(Johansson)

 6 Nov

Introduction to neuro-symbolic AI from a cognitive science perspective

(Dubhashi)
2 10 Nov

Introduction to program synthesis

(Johansson)

 13 Nov

Intro to Reinforcement Learning

(Dubhashi)
3 17 Nov

Introduction to neuro-symbolic AI for mathematics and reasoning

(Johansson)

 20 Nov

From Chain-of-thought to reasoning models

(Dubhashi)
4 24 Nov

Curriculum learning

Guest lecture: Stefano Sarao Manelli, Chalmers

 27 Nov

AI for math continued: From AlphaProof to agents solving Milennium problems.

(Johansson)
5 1 Dec

Symbolic learning, DSLs and the ARC challenge

Guest lecture: David Cerna, NDEA

 6 Dec TBA
6 8 Dec TBA 11 Dec TBA
7 15 Dec

Neuro-symbolic AI for science

(Maryam Dadkhah & Guest lecturer Knut Andreas Meyer, Chalmers)

 18 Dec TBA

Office Hours with TAs: Getting help with the Assignment

If you want to get help with the assignment you may come to the Office Hours and speak to the TA. 

Office hours dates:

  • Monday, 7 Dec: 11:00-12:00. CSE EDIT 5128.
  • Friday, 11 Dec: 14:00-15:00. CSE EDIT 4128.

If you need help outside office hours, please contact either TA via email and we can organise a meeting.

 

Course literature

The course literature consist of recent articles from the research literature. Links to reading material related to each lecture will be linked to from the respective lecture page, clearly stating which are obligatory readings and which are supplementary for the interested. After the lecture, you will also find copies of the slides there. You should be prepared to read quite a lot each week for this course, including scientific articles. 

Course design

The course has two lectures per week. The lectures will introduce topics from both symbolic AI, machine learning and neuro-symbolic AI. Students are expected to attend lectures and to read the material linked from each lecture page. To contact the teachers, please use email. There will likely be a delay in getting a reply to messages sent in Canvas.

Optional Assignments:

There will be one hand-in programming assignment and a presentation of a research paper. These should be done in pairs and the programming assignments should be submitted via Canvas by the deadline stated. They are not strictly mandatory, but give bonus points towards the final grade (see below under Examination for details). Assignments handed in before the deadline will be graded. There will be no second submissions. Deadlines are firm, unless granted an extension by the examiner for valid reasons such as illness. Vacation trips, failure to read the deadline dates etc does not count as valid reasons. It is the student's responsibility to keep themselves informed about the course by reading updates and announcements on the course Canvas page frequently.

Dates for the assignments are:

Week Assignment Deadline/Presentation
5 Select Research Paper for Presentation Fri, 4 Dec (at 23:59)
5-6 Lab Assignment Tue, 15 Dec (at 23:59)
7 Research Paper Presentations (Session 1) Wed, 16 Dec (8:00-10:00)
7 Research Paper Presentations (Session 2) Fri, 18 Dec (8:00-10:00)

Changes made since the last occasion

The following changes have been made since the last occasion (HT25):

  • The lectures have been updated to reflect new research in the field of neuro-symbolic AI.
  • The programming assignment is new this year, and is a single larger one instead of several small ones.
  • The exam is digitalised in Inspira. As such, the total number of points, bonus points and grade limits has been slightly adjusted (see under Examination).

Learning objectives and syllabus

Learning objectives:

After completion of the course the student should be able to:

  • Separate what characterise symbolic and neural AI.
  • Explain what neuro-symbolic AI encompasses.
  • Apply and implement methods and algorithms for neural AI
  • Apply and implement methods algorithms for symbolic AI.
  • Apply and implement neuro-symbolic AI methods and algorithms.
  • Read and understand recent research articles in neuro-symbolic AI.

Link to the syllabus on Studieportalen.

https://www.chalmers.se/en/education/your-studies/find-course-and-programme-syllabi/course-syllabus/DAT615/?acYear=2026/2027

Examination form

The course has a written exam using the Inspera digital examination environment. Please download the appropriate software well ahead of the exam (refer to Chalmers student webpages about examination for details, dates and registration procedures).

The regular exam is in January and potential re-exams in the spring and in August (note that due to the very low number of students taking the re-exams, we strongly encourage notification also to the examiner stating which re-exam you intend to take, there has been cases where few or no students register in which case time is better spent elsewhere than in constructing exam questions).

No aids are allowed. The exam is compulsory. Please note that past exam papers are not provided as the examination is multiple choice. Students are strongly encouraged to read and engage with the course material when studying, and not rely on past exam papers. This will give you a better foundational understanding, encourages depth rather than shallow memorisation and avoids tricking yourself into overfitting to past questions. Note also that as the course aims to follow the research frontier, there will be some variation from year to year.

Examination format

The exam has two parts.

  • Part A consists of 30 multiple choice questions (1 mark each) covering the full range of topics from the course, including guest lectures. To proceed to part B, students must pass Part A. If you pass Part A, you are guaranteed at least grade 3. If you are satisfied with this, you do not need to proceed to Part B. As the exam is digital, you will see your score immediately (but not the correct answers).
  • Part B consists of two longer-answer questions of a problem-solving nature (total 20 marks), but will not require programming, complicated calculations or anything like that which requires additional tools. To have a chance on grade 4 or 5, you will also need to do Part B.  

Bonus points from assignments [THIS PARAGRAPH IS UNDER CONSTRUCTION - REGARD AS DRAFT ONLY]

The programming assignments, paper presentation and attendance of another presentation, where you act as an opponent, count for a maximum of 3 bonus points:

  • The programming assignment gives max 1 bonus point, for passing the assignment.
  • The research paper presentation gives max 2 bonus points, based on the following criteria (judged by the teacher):
      • You must demonstrate that you understand the main research question and results of the presented paper.
      • You must present the material clearly and using appropriate audio-visual aids.
      • Your answers to the opponent's questions must demonstrate a good understanding of the paper.
      • As the opponent, you must ask insightful questions about the presentation, demonstrating that you understood that other paper as well.

Course summary:

Course Summary
Date Details Due