COMP 5970/6970: Federated Learning

Instructor: Jiaqi Wang
Lecture: M 15:00-17:30
Location: Shelby Center 1120
Office Hours: M 17:30-18:30
TA: Huy Hung Nguyen, hhn0008@auburn.edu

Course Overview

This course explores advanced techniques in federated learning, collaborative machine learning, and multi-agent systems that enable decentralized model training across distributed data sources. Also, we will examine recent developments in deploying large language models (LLMs) and foundation models in federated and collaborative settings, with a focus on real-world applications in computer vision, Internet of Things (IoT), and healthcare. Students will engage with state-of-the-art literature, identify open research problems, and critically analyze recent advances shaping the future of collaborative intelligence.

Schedule

The schedule is tentative and subject to change. Please see the table below for the latest updates.

Date	Topics	Dissert	Notes	Due
01/12	Course Overview		Syllabus and logistics
01/19	No Class		M.L. King, Jr. Day
01/26	Foundation of Collaborative Machine Learning	How to read papers
02/02	Heterogeneity in Federated Learning	How to get intenship/full-time jobs	1st Batch Paper Released
02/09	Federated Semi-supervised Learning	How to conduct professional presentation
02/16	Paper Presentation
02/23	Paper Presentation			Paper slides due
03/02	Federated Learning Meets Large Models	Federated learning implementation	2nd Batch Paper Released	1st bonus idea due
03/09	Project Pitch Presentation
03/16	Project Pitch Presentation			Pitch slides due
03/23	Multi-agent Systems	How to write and sumbit papers		Proposal writing due
03/30	Guest Speaker		TBD, industry or academia
04/06	Paper Presentation			2nd bonus idea due
04/13	Paper Presentation			Paper slides due
04/20	Final Presentation
04/27	Final Presentation			Slides due
05/01(F)	No Class			Project report due

Grading

Paper Presentation and Slides (20%):
- There are two paper presentations. Each student is required to present one paper in each round.
- Each slides and each presentation count towards 5% of the final grade.
- Online students need to share the link of the video recording for the presentation.
Project Proposal (30%)
- Project proposal writing report counts towards 10% of the final grade.
- Project proposal pitch presentation counts towards 20% of the final grade.
- Online students need to share the link of the video recording for the presentation.
Final Project (50%)
- Project final writing report counts towards 30% of the final grade.
- Project final presentation counts towards 20% of the final grade.
- Online students need to share the link of the video recording for the presentation.
Bonus
- Idea bonus: Each submission is 1-page of what you may think is interesting or any bold reseach ideas based on the lectures. Each submission adds 5 points to the actural grade. We totally have 2 submission opportunities.
- Paper submission: You are encouraged to submit your project paper to AI/ML/DM conferences. The instructor will provide a comprehensive evaluation of the work before the final grading day. The submission is required to be reviewed and approved by the instructor. The approved paper adds 10 points to the actural grade. Please notice that the bar of the paper submission could be (much) higher than a course project.
- Course survey: The survey completion rate >80% leads to an additional 10% for everyone’s actual grades, i.e., your_final_grade = your_actual_grade * 110%

Late Policy

Late assignments are generally not accepted. If you are unable to meet a deadline due to extenuating circumstances, you must notify the instructor as early as possible in advance.
May be submitted late with a 25% penalty for every 12 hours late, up to 48 hours (2 days). After 48 hours, no credit will be given for the submission.

Academic Integrity

Auburn University is dedicated to honesty and strong moral behavior in academics. Cheating and plagiarism are expressly prohibited by the Auburn University Academic Honesty Code. Students who attend Auburn are expected to attain high competency and deep understanding in their areas of study. While developing skills and knowledge, it is essential that Auburn students commit themselves to core principles and behaviors consistent with academic and personal integrity:

Honesty: Upholding trust and honesty by doing your own academic work and not cheating.
Fairness: Following correct academic procedures and practices as stated in course guidelines and as defined by Auburn University.
Respect: Growing as a student by facing academic challenges and interacting productively with instructors.
Responsibility: Being accountable for and accepting responsibility for class assignments and personal academic development.

More details can be found in the Auburn University Academic Honesty Code.

Instructor's extra notes: we have zero tolerance for academic dishonesty. If you are cheating in any forms (slides, presentations, writing etc.), you will receive a zero for the assignment and may fail the course. I hope we can trust and respect each other.

Accommodation Policy

Our lectures strictly follow the Auburn University policies on accommodations for students. If you have a disability and require accommodations, please contact the Office of Accessibility at Auburn University. Also, please discuss your accommodations and needs with the instructor as early as possible before the course begins. I will work with you to ensure that accommodations are provided appropriately.

Syllabus and Grading Change Policy

The Instructor reserves the right to make changes to the syllabus and grading policy as necessary. Any changes will be communicated to students in a timely manner.