Machine Learning for Artificial Intelligence

Code

IT-MAL2

Version

1.1

Offered by

ICT Engineering

ECTS

Prerequisites

Knowledge about machine learning corresponding to MAL1 or similar.”

Main purpose

"Machine Learning for Artificial Intelligence" is a course that explores the fundamental concepts, techniques, and applications of deep learning in the context of artificial intelligence (AI). This course is designed to provide students with a comprehensive understanding of how deep learning methods can be leveraged to solve complex AI problems.

Knowledge

After having completed the course, the student will have gained knowledge about algorithms, methods, techniques, tools, and applications within the following:

- Different types of neural networks, e.g. feed-forward, convolutional and recurrent neural networks.
- Predictive methods, e.g., image classification and speech recognition.
- Generative methods, e.g., generative adversarial networks (GANs) and generative pre-trained transformers (GPTs).
- Reinforcement methods, e.g., game AI.

Skills

Upon completion of this course, students should be able to:

- Explain and apply a range of deep learning methods for AI.
- Implement and fine-tune deep learning models in a programming language.
- Apply ethical considerations when developing AI systems.

Competences

Upon completion of this course, the goal is that the students have acquired the competences to:

- Make informed choices about the use of deep learning methods.
- Communicate and discuss the theory, tools and techniques of deep learning and artificial intelligence.
- Discuss, address and reflect upon ethical aspects of using artificial intelligence.

Topics

Teaching methods and study activities

The mode of teaching will be classroom based and will involve lectures by the teacher and exercises/assignments made in class. The students are expected to participate actively in the lectures and to work on the course between classes.

During the semester, the students should solve a number of assignments. These will make up their portfolio. In addition to this, the course concludes with a larger group project. These will form the basis of the exam.

The total work-load for the student is expected to be around 125 hours.

Resources

Literature:

Géron, A. (2022). Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems (3rd ed.). O’Reilly.

Other:

Additional material will be uploaded to Itslearning.

Evaluation

Examination

Exam prerequisites:
None.

Type of exam:
Individual oral exam, 20 minutes without preparation.
At the exam, the student will randomly draw one of the portfolio assignments. The exam will then take place as a discussion of this assignment, the students’ group project and the curriculum in general.
The final grade will be based on an overall assessment of the assignments, the group project and the oral examination.
Internal assessment.

Tools allowed:
The student is expected to bring their portfolio assignments and their final project to the oral exam, such that they are able to display and run their code.

Re-exam:
The re-exam is the same as the ordinary exam.

Grading criteria

Grading based on the Danish 7-point scale.

Additional information

Responsible

Frederik Thorning Bjørn (frbj)

Valid from

01-02-2025 00:00

Course type

Keywords

Artificial intelligence. Deep learning. Neural networks. Image classification. Speech recognition. Natural language processing. Reinforcement learning.