Official course description:
AbstractThe overall goal of the course is to introduce students to a selection of the most important problem solving and decision support techniques within AI and optimization.
Computers are increasingly used to support decision making, exploiting that computers can perform computations at scale, on large amounts of data, requiring large amounts of computation steps, or both. This course studies how to solve such decision problems
The goal is to make students able to identify, design, and implement efficient solutions to the kind of decision problems that arise in modern organizations and IT products. The expectation is that a student mastering the material is able to work in internationally leveled business intelligence and optimization groups as well as in development departments of “intelligent” applications as used in smart phones, computer games, enterprise resource planning systems, and decisions support systems.
The course will cover the followings topics:
- Informed search: greedy heuristic search, A*, breadth-first heuristic search
- Local search: hill-climbing, simulated annealing, genetic algorithms, tabu search, population-based search
- Adversarial search: Minimax search, alpha-beta pruning
- Markov Decision Processes
- Representations: disjunctive (DNF), conjunctive (CNF), and if-then-else (INF) normal forms, Binary Decision Diagrams (BDDs)
- Reasoning: resolution, SAT-checking
- Local consistency: arc-consistency, path-consistency, i-consistency
- Look-ahead search strategies: forward-checking, arc-consistency look-ahead, maintaining arc-consistency
- Simplex algorithm
- You must have passed an elementary programming course. (for example
- Introductory Programming)
- You must have passed a discrete mathematics course (for example Foundations of Computing, Discrete Mathematics)
- You must follow in parallel, or have passed an introductory algorithms course (for example Algorithms and Data Structures).
If you are an external student, it is important that you have programming-experience from elsewhere, i.e. through a daily use in a developer position in the software industry.
Intended learning outcomes
After the course, the student should be able to:
- Identify decision problems in work processes and IT products that can be solved by AI and optimization algorithms.
- Implement AI and optimization software components to solve these problems efficiently.
- Apply standard AI and optimization models and solvers.
- Participate in concept development of advanced decision support systems.
Learning activities consist of lectures and exercises
- Mandatory Exercises. You must hand-in and pass 3 out of 10 exercises to qualify for taking the exam. You can start working on a mandatory problem, as soon as the exercise each week is posted in LearnIT. This will at the latest happen on the day of the recitation of the exercise. You hand in a mandatory problem at the next recitation session. This means that you have at least one week to work on the exercise. You can work on mandatory problems in groups, but they must be handed in individually. The teaching assistants will spend one week evaluating the exercises and they will be handed back at the next recitation session. Students who do not pass 3 exercises will get a possibility to turn in revised solutions before the exam. Details will be available in LearnIT
- Mandatory projects. During the semester there will be 3 implementation projects. You must hand-in and pass 2 of these to qualify for taking the exam. The projects are made in groups of 2-3 students. You approximately 3 weeks to finish each project. Students who do not pass 2 projects will get a possibility to turn in revised solutions before the exam. Details will be available in LearnIT.
- The pedagogical function of the mandatory activities is to provide a setting for learning where the students can practice specific ILO’s (e.g. Participate in concept development of advanced decision support systems).
- After each hand-in and project, the students will be given formative feedback from the teacher/TA/peer in order to scaffold their conceptual understanding of the given subject
The student will receive the grade NA (not approved) at the ordinary exam, if the mandatory activities are not approved and the student will use an exam attempt.
The course literature is published in the course page in LearnIT.
Student Activity BudgetEstimated distribution of learning activities for the typical student
- Preparation for lectures and exercises: 20%
- Lectures: 20%
- Exercises: 20%
- Assignments: 15%
- Project work, supervision included: 15%
- Exam with preparation: 10%
Ordinary examExam type:
A: Written exam on premises, External (7-point scale)
A22: Written exam on premises with restrictions.
Restricted access - LearnIT only
Written and printed books and notes
E-books and/or other electronic devices
B: Oral exam, External (7-point scale)
B22: Oral exam with no time for preparation.