Course content

The course *Cognitive Systems* aims to deepen understanding of the diverse cognitive capabilities and functions found in living beings, particularly humans, which set a benchmark for expectations in artificial intelligent systems. The first third of the course focuses on exploring the nature of cognition, while the following sections address the challenges of implementing human-like cognition in technical systems. The course offers a comprehensive overview of the historical development of cognitive theories and technologies, from both technical, philosophical, and psychological perspectives, and proceeds towards modern approaches in AI.

The course introduces various approaches to realizing cognitive capabilities in cognitive architectures, covering both classical and contemporary systems. It emphasizes the broad range of cognitive abilities and the unique challenges these present for artificial systems. Key topics include different modalities of reasoning and learning, perception, and planning, ultimately leading to a critical analysis of current claims about the level of intelligence in state-of-the-art AI architectures.

Learning outcome

Knowledge: This course provides a comprehensive introduction to key concepts, theories, and experimental findings in human cognition and artificial intelligence. Initially, we explore foundational cognitive science, covering core philosophical and psychological theories, before progressing to artificial cognitive systems. The course examines computational theories of mind and their implementation in selected cognitive architectures, including both classical symbolic AI approaches as well sd modern techniques. Topics include reasoning, perception, fuzzy inference, non-deductive and probabilistic reasoning, as well as selected aspects of machine learning and artificial neural networks.

Skills: Upon completing this course, students will be able to apply and critically compare various methods and approaches to cognition in artificial intelligence. They will gain practical experience in analyzing and evaluating cognitive models and systems relevant to AI applications.

General Competence: Through lectures, tutorials, and the study of seminal publications, students will develop the competence to engage in informed discussions, evaluate diverse AI methodologies, and make decisions within the field of intelligent systems. Practical exercises and theoretical assignments will equip students with the skills to contribute meaningfully to research and development in AI and cognitive science. Students will also gain a historical perspective on AI and cognitive science, including selected aspects of the philosophy of mind, and practical applications in embodied AI.

Learning methods and activities

A: Lectures, B: Tutorial hours / colloquia where student questions are discussed, the lecture topics are extended. Exam preparations supported by self-test question or task sets. C: Self-study / reading, and homework assignments.

A number of mandatory exercises must be approved in order to take the midterm and final exams.

Compulsory assignments

• Exercises
• Exercises

Recommended previous knowledge

TDT4136 Intro to AI and the 1st part of the new TDT4172 Intro to Machine Learning course. Both recommended, but not mandatory.

Course materials

To be announced