The following fields are treated within the scope of the course:

• Fundamental AI problems and solutions including search algorithms and planning, knowledge representation forms and knowledge including reasoning strategies, decision support and heuristics.
• Intelligent agents and multi-agent systems
• Automatic analysis and generation of natural language.
• Machine learning and neural networks.

Focus is on artificial intelligence for knowledge-based systems, agent system and strategies.

After passing the course, the students should be able to:

• give an account of artificial intelligence and its application areas
• know and account for artificial intelligence methods and technologies
• formulate and carry out a well delimited and qualified assignment that applies artificial intelligence techniques.

To learn the different AI areas the course is divided into lectures, tasks, seminars, laboratories, and quizzes.

Lectures

The online recorded lectures are introduced providing a summary of the recorded lectures that are provided via Canvas.

The summary lectures together with the recorded lectures give a complete view of the different areas. The summary lecture only provides an insight into a particular area, whereas the recorded lectures provide more details. More specifically:

The lectures present the following topics:

• Introduction to AI, vision, and future
• Searching, planning, and scheduling
• Decision-support system, expert systems, and knowledge-based systems including reasoning strategies and heuristics, software development
• Software agents, intelligent agents, and multi-agent systems
• Machine learning
• Neural networks and deep learning
• Natural language processing

Tasks (Pass / Fail)

For each AI topic, see lectures above, there is a task of writing a short description of the AI area and providing five questions together with answers.

Seminars

The seminars follow the lectures. During the seminars, the current AI area is discussed with techniques and problems. Also, the questions and answers, that the students have provided for the task of the AI area, are discussed. The lectures, seminars with questions, answers, and tasks shall prepare the students for the quizzes. For tasks and quizzes, see below.

Laboratories (Pass / Fail)

The laboratory consists of practical programming tasks. These tasks are for imparting theoretical knowledge.

Quizzes (Pass / Fail)

Online quizzes are provided for each topic. The two quizzed contain:

1) closed-questions with answer alternatives and

2) one open question.

The open question can only be answered after the closed questions are correctly answered. The number of tests for the 1) closed-questions with answer alternatives is unlimited whereas the 2) one open question can only be answered once.

• Correct answers on both quizzes lead to a pass on this part.
• A failure in any of the quizzes lead to fail on this part and requires to retake this part during the written exam.*)

*) To clarify this part: the written exam includes one part where questions about the different AI topics are asked (that is the AI topics presented during the course), see the topics in the lecture section above.

The student only needs to take the questions for the AI topics that failed during the course. When this is done, the student will receive the grade E. However, to get a higher grade the second part of the written exam must be taken, for more information see the written exam below.

Written Exam (A-F)

The written exam includes two parts:

• the first part with all the AI topics that are presented during the course, see lectures above. The part is graded with a pass or fail and gives an E on the written exam. Observe that this part only needs to be taken, if and only if, any, some, or all of the quizzes for the different AI tasks failed. The students only take the questions that correspond to the failed quizzes. 
• the second part is about an AI problem to be solved. The solution is to provide and, in detail, describe the select AI technique and tool(s), that must be correct for solving the particular AI problem.

Schema HT-2021-1005

HÅKANSSON A., HARTUNG, R.L., 2020: ARTIFICIAL INTELLIGENCE Concepts, Areas, Techniques and Applications , Studentlitteratur, Lund. (main literature)

NORVIG, P., RUSSEL, S., 2021: Artificial Intelligence: A Modern Approach. Pearson Education Limited; 4th edition. (complementing book)

+ Articles and papers, and other recommended literature

Own laptop

Programming language, well-known to the student, such as:

Python

Java

Prolog

But also other programming tools, such as:

Other programming languages, tools and libraries, for example:

VisiRule

GAMA

WEKA

Tensorflow

Keras

Students at KTH with a permanent disability can get support during studies from Funka:

Funka - compensatory support for students with disabilities

A, B, C, D, E, FX, F

• INL1 - Written assignment, 4.0 credits, Grading scale: P, F
• TEN1 - Examination, 3.5 credits, Grading scale: A, B, C, D, E, FX, F

Based on recommendation from KTH’s coordinator for disabilities, the examiner will decide how to adapt an examination for students with documented disability.

The examiner may apply another examination format when re-examining individual students.

Written examination. Written assignment that is reported in groups.

Written assignment ( INL1 )

Individual tasks for the different AI-topics presented during the course, see lectures. This is graded with pass/fail.

Labs for each AI topic, individually or in a student group of two students. The students must demonstrate a well-functioning program. This is graded with pass/fail.

Two-students main project solving a particular AI problem. This task includes a written report describing the problem and solution, and well-functioning software.

Examination ( TEN1 )

There are two quizzes for each AI topic. Students answer these quizzes, individually. Correct answers to all questions on all quizzes for all AI topics results in the grade E on TEN1.

A written exam with two parts (can improve the grade on the course, from E, see quizzes above):

• one part for the AI-topcis where the students that have failed the quizzes must take, graded pass / fail.
• second part with a particular AI problem to be solved, grade A-F.

Individual tasks, labs, individual quizzes, and the main project:

All tasks (task 1 - task 7) are graded with a pass or fail.

All labs are graded with a pass or fail.

All quizzes are graded with a pass or fail.

The main project is graded with a pass or fail. However, an excellent project with high-quality programming solving an very difficult task can increase the final grade on the course. Hence, students that are between two grades can get a higher grade thanks to the main project.

If any or all of the quizzes are graded with fail, the students can take the written exam, with the two parts.

A second written exam can be taken, that can increase or decreas the final grade.

Missed any of the tasks, labs and two-part quizzes for each AI-topics:

- one part of the written exam is about the AI-topics. The students that have failed any of the quizzes during the course, can take the questions that concern the AI-topic(s) that the student failed earlier.

Failed tasks and labs must be corrected and handed in via Canvas.

All parts need to be finalized before the course ends. Activities outside the course frame will not be corrected.

Results are in Canvas and Ladok.

• All members of a group are responsible for the group's work.
• In any assessment, every student shall honestly disclose any help received and sources used.
• In an oral assessment, every student shall be able to present and answer questions about the entire assignment and solution.