CNG462 ARTIFICIAL INTELLIGENCE
| Course Code: | 3550462 |
| METU Credit (Theoretical-Laboratory hours/week): | 3 (3.00 - 0.00) |
| ECTS Credit: | 6.0 |
| Department: | Computer Engineering |
| Language of Instruction: | English |
| Level of Study: | Undergraduate |
| Course Coordinator: | Assist.Prof.Dr MARIEM HMILA |
| Offered Semester: | Fall Semesters. |
Course Objectives
This course is an introduction to the fundamental problems of AI and general approaches employed by the AI community. We will cover knowledge representation, heuristic search, problem solving, game-playing, logic and deduction, planning, machine learning and natural-language processing.
After taking this course, students will have a basic understanding of major AI tools and algorithms. They will be able to choose suitable tools from the AI toolbox to solve a particular problem and be able to adjust the generic algorithm to fit a particular purpose. This course will provide necessary skills to solve constrained optimization problems, solution modeling under input errors, and classification problems.
At the end of this course, students will be able to:
- Express data and knowledge by using logical models and knowledge representation schemes (PI-a2).
- Apply logical principles for sound reasoning to prove theorems (PI-a3 ,e1).
- Analyze knowledge rich domains to formulate heuristics (PI-c3).
- Understand algorithms for search, adversary search, CSP and planning to develop efficient solutions (PI-c3, e1).
- Analyze real-life applications to represent uncertain knowledge with the help of probability theory (PI-a6).
- Understand algorithms for probabilistic reasoning in Bayesian Networks (PI-a6).
- Understand algorithms for simple and sequential decision making (PI-a6, e1).
- Understand requirements for supervised, unsupervised and reinforcement learning and apply machine learning algorithms (PI-a6, c3).
Course Content
Basic LISP programming; picture analysis WALTZ algorithm; game playing, game trees, the mini-max rule, alpha-beta pruning technique; nature language understanding, transformation of grammar, ATN grammars, techniques used in semantics.
Course Learning Outcomes
Relationship of Course to Student Outcomes:
Satisfies the following student outcomes (SOs) via the following Performance Indicators:
- SO-1: PI-a2.
- Express data and knowledge by using logical models and knowledge representation schemes.
- SO-1: PI-a3, SO-1: PI-e1.
- Apply logical principles for sound reasoning to prove theorems.
- SO-2: PI-c3.
- Analyze knowledge rich domains to formulate heuristics.
- SO-1: PI-e1, SO-2: PI-c3
- Understand algorithms for search, adversary search, CSP and planning to develop efficient solutions
- SO-1: PI-a6.
- Analyze real-life applications to represent uncertain knowledge with the help of probability theory.
- SO-1: PI-a6, SO-1: PI-e1.
- Understand algorithms for simple and sequential decision making.
- SO-1 – PI-a6, SO-2: PI-c3.
- Understand requirements for supervised, unsupervised and reinforcement learning and apply machine learning algorithms.
Program Outcomes Matrix
| Level of Contribution | |||||
| # | Program Outcomes | 0 | 1 | 2 | 3 |
| 1 | Employ knowledge of mathematics, science and engineering to formulate solution to real life computing problems | ✔ | |||
| 2 | Design and conduct experiments, as well as analyze, evaluate and interpret data | ✔ | |||
| 3 | Design systems, components, and/or processes by specifying the requirements and determining the realistic constraints such as ethical and environmental | ✔ | |||
| 4 | Judge professional and ethical principles and integrate them in the working environment | ✔ | |||
| 5 | Have the ability to communicate effectively | ✔ | |||
| 6 | Recognize the need for, and an ability to engage in life-long learning | ✔ | |||
0: No Contribution 1: Little Contribution 2: Partial Contribution 3: Full Contribution