TSKS38 · 6 credits · Advanced level (A1N) · Elektroteknik
Distributed Information Processing and Machine Learning
Active · next start HT 2026
47%
passed the latest ordinary exam
TEN1 · 15 Jan 2026 · 17 students
47%
passed the ordinary exams, all sittings over the last 5 years
TEN1 · 17 students
Part of 5 programmes
Exam
TEN1 Written examination
- 5
- 4
- 3
- U
Ordinary exams
47% passed · 17 students
- 47%17
Re-exams
80% passed · 5 students
- 80%5
- Pass the re-exam: 80% (5 students), compared with 47% (17 students) on the ordinary exam
- Get 4 or 5: 32% (22 students)
Show small sittings (1)
Sittings with fewer than 5 students, often credit transfers or special sittings. They are not part of any figure on this page.
- 50%2
Other modules
Partial exams, hand-ins and labs. They are not part of the figures at the top.
Hand-in or lab
LAB1 Laboratory work
- G
Main rounds
100% passed · 13 students
- 100%13
Later rounds
No sittings with at least 5 students.
Show small sittings (2)
Sittings with fewer than 5 students, often credit transfers or special sittings. They are not part of any figure on this page.
- 100%1
- 100%3
About the course
Aim
After completing this course, the student is expected to:
- explain and summarize theoretical methods, perform mathematical derivations, and analyze specific technical concepts related to distributed consensus, optimization, and machine learning over networks.
- implement and execute laboratory tasks using a standard programming language, verify and reflect on the theoretical results related to consensus formation and collaborative machine learning through simulation examples.
- present and report laboratory work in a structured and coherent manner.
Content
- Basics about communication networks, topology and architecture, graphs and spectral graph theory.
- Examples of signal and information processing over networks. Distributed control, inference, and machine learning applications.
- Information dissemination and fusion over networks. Consensus protocols and gossip algorithms. Distributed average consensus in time-invariant and time-variant networks.
- Optimization methods for machine learning. Gradient descent, stochastic gradient descent, distributed and decentralized gradient descent.
- Centralized and decentralized collaborative machine learning.
- Federated learning. Communication efficiency, data heterogeneity and privacy.
- Decentralized learning. Communication-computation tradeoffs.
- Sustainable AI: ethical aspects of AI, energy efficiency, computational efficient ML etc.
- Examiner
- Zheng Chen
- Scheduled hours
- 52 h
- Main field of study
- Elektroteknik