Teorimoduler - IDI

TDT05 - Developing ML Tools - Insights and Case Studies


Prof. Zhirong Yang

Course content

You may have known how to build a machine-learning application using existing tools. Are you curious about how these tools are developed from scratch? If you're driven to create an innovative ML tool that could redefine the landscape, this course is tailored for you.

Crafting a pioneering machine learning tool is a task that blends brilliance with innovation, demanding not only a strong foundation in mathematics and programming but also a deep well of inspiration. In this course, we will review and discuss the development clue of some selected milestone algorithms and models. By exploring these success stories, you will gain valuable insights into making strategic choices in direction, formulating problems, fitting models, and executing software implementations.

Course grading

  • Presentation (70%)
  • Opponent and discussion (10%)
  • A position paper or a small group project (20%)
  • No exam


Attendance is mandatory! Classroom and time are TBD

Office hours (Tuesdays 10-12)

  • Zhirong Yang, Gløshaugen IT-Bygget 320

Recommended prerequisites

  • University mathematics
  • Programming (We mainly use Python; some other languages are possible)
  • Machine learning basics


  • Register for the course here.
  • CS students have to register for a project and get approval from their supervisor to access the theory modules.
  • Non-CS students can choose "Informatics" in the dropdown.
  • For students who are not in computer science, please consult your own department to see whether they have the same format of specialization courses. Otherwise, we may not grant credits to you.

Microsoft Teams

  • We will mainly use Microsoft Teams for the course materials
  • We also use Piazza for discussions and Q&A
4217 Visninger