OHJ-2556 Artificial Intelligence, 6 cr

Prof. Tapio Elomaa

Spring 2011, Jan. 11 - May, 5

Lectures: Tue 12-14 TB219 and Thu 12-14 TB223

Exercise group: Tue 16-18 TC163 M.Sc. Teemu Heinimäki

Programming assignment: M.Sc. Timo Aho

Course exam: Wed. May 11, 2011, 9-12 AM

News

The Essay deadline has been extended until MAY 22, 2011
The programming assignment has been published. See here. The deadline is APRIL 17, 2011.
Essay instructions are now available, the deadline is MAY 22, 2011.

General Information

A course on programming (e.g., OHJ-1156 Programming II) is an obligatory prerequisite.
Courses OHJ-2156 Analysis of Algorithms and OHJ-2306 Introduction to Theoretical Computer Science help to follow the lectures and are, thus, recommended prerequisites.
The components of the course are:
- Lectures (attend!, optional)
- Weekly exercises (attend!, yields max 6 extra points)
- Essay (compulsory, max 10 points), Deadline May 22, 2011
- Presentation on the essay topic (yields max 4 extra points)
- Programming project (yields max 6 extra points)
- Exam (compulsory)

Lectures

The lectures are based on the textbook:

Russell & Norvig: Artificial Intelligence: A Modern Approach, Third ed., Pearson, 2010.

Week	Dates	Slide Numbers	Chapters in the Course Book
1	Jan. 11 and 13	1 - 24	1 Introduction, 2 Intelligent Agents
2	Jan. 18 and 20	25 - 58	3 Solving Problems by Searching
3	Jan. 25 and 27	59 - 77	3 & 4 Beyond Classical Search
4	Feb. 1 and 3	78 - 108	4 & 5 Adversarial Search
5	Feb. 8 and 10	109 - 136	7 Logical Agents
6	Feb. 15 and 17	137 - 165	7, 8 First-Order Logic, 9 Inference in First-Order Logic
7	Feb. 22 and 24	166 - 190	9, 13 Quantifying Uncertainty
8	Mar. 8 and 10	191 - 209	13
9	Mar. 15 and 17	210 - 240	14 Probabilistic Reasoning
10	Mar. 22 and 24	241 - 266	14, 16 Making Simple Decisions
11	Mar. 29 and 31	267 - 282	17 Making Complex Decisions
		Guest Lecture on Watson
12	Apr. 5 and 7	283 - 302	17, 18 Learning from Examples
13	Apr. 12 and 14	303 - 330	18, 20 Learning Probabilistic Models
14	Apr. 19 and 28	ANN and SVM	18.7, 18.9
		Student presentations on Thu Apr. 28 Sindhuja Ranganathan: Data stream mining Hong Liu: Adaptive data stream mining Georgy Minaev: Artificial neural nerwork - Artificial life
15		Student presentations on Tue May 3 Arturo García: Emotions in AI Kira Neubehler: Requirements for natural language processing Martin Fábry: Multiagent systems Student presentation on Thu May 5 Abhishekh Gupta, D. Veerendra Kumar: Clustering algorithms

Weekly Exercises

There is just one exercise session per week. It is concerned with topics that were lectured the previous week. Questions to next week's exercises are published on this page on previous Wednesday.

Exercise	Problems
1	Jan. 25, 2011
2	Feb. 1, 2011
3	Feb. 8, 2011
4	Feb. 15, 2011
5	Feb. 22, 2011
6	Mar. 8, 2011
7	Mar. 15, 2011
8	Mar. 22, 2011
9	Mar. 29, 2011
10	Apr. 5, 2011
11	Apr. 12, 2011
12	Apr. 19, 2011

Home Work

The programming assignment has been published. See here. The deadline is APRIL 17, 2011.
The game that will be considered in the programming project is Afrikan tähti (the star of Africa)
Each student (in groups of 1-3, may choose freely a topic within the scope of the course and write an essay of 10-15 pages concerning the topic.
Sample topics include Decision rule learning (Background, CN2, Ripper), Decision tree learning (C4.5 and forests), Ensemble learning, Modern applications of AI, Constraint satisfaction problems (reaching beyond the textbook), Frequent pattern mining, Weak and strong AI, Ontological engineering and semantic web, Natural language processing, Visual perception, Robotics (and particular topics within it), robot navigation, multiagent systems, Condenced representations for frequent sets, Support vector machines, Genetic algorithms, Neural networks, Planning algorithms, Probabilistic reasoning over time, Philosophy of artificial intelligence, Clustering algorithms, The EM algorithm, Unsupervised learning, Semi-supervised learning, Data stream mining, Chess playing algorithms, Practicalapplications of reinforcement learning, Random forests, Regression trees, Grammar induction, Probabilistic language processing, Emotions in AI, Document topic modeling algorithms, Instance-based learning, Speech recognition, Face recognition.
You can modify the sample topics to suit your own interests, or if you have a specific application that you are interested in (a game for example), you may consider the application of AI methods to it. If you are uncertain about the suitability of your own topic, please contact Teemu Heinimäki for approval. A short presentation on the essay may be given in late April, early May.
Here are the instructions for the essay. Please deliver your essay to M.Sc. Teemu Heinimäki preferably as a pdf file.

Topics

We intend to cover the following chapters from the textbook:

Lecture week Topic

1 I Artificial Intelligence

1 Introduction

2 Intelligent Agents

2-4 II Problem-solving

3 Solving Problems by Searching

4 Beyond Classical Search

5 Adversarial Search

5-7 III Knowledge, reasoning, and planning

7 Logical Agents

8 First-Order Logic

9 Inference in First-Order Logic

10 Classical Planning

11 Planning and Acting in the Real World

8-10 IV Uncertain knowledge and reasoning

13 Quantifying Uncertainty

14 Probabilistic Reasoning

16 Making Simple Decisions

17 Making Complex Decisions

11-14 V Learning

18 Learning from Examples

19 Knowledge in Learning

20 Learning Probabilistic Models

21 Reinforcement Learning

Lecture week	Topic
1	I Artificial Intelligence
	1 Introduction
	2 Intelligent Agents
2-4	II Problem-solving
	3 Solving Problems by Searching
	4 Beyond Classical Search
	5 Adversarial Search
5-7	III Knowledge, reasoning, and planning
	7 Logical Agents
	8 First-Order Logic
	9 Inference in First-Order Logic
	10 Classical Planning
	11 Planning and Acting in the Real World
8-10	IV Uncertain knowledge and reasoning
	13 Quantifying Uncertainty
	14 Probabilistic Reasoning
	16 Making Simple Decisions
	17 Making Complex Decisions
11-14	V Learning
	18 Learning from Examples
	19 Knowledge in Learning
	20 Learning Probabilistic Models
	21 Reinforcement Learning

Course Grading

Compulsory parts: Exam (May 11, 2011) yields up to 30 points and the essay (deadline May 18, 2011) up to 10 points
Voluntary parts: Programming assignment (deadline April 17, 2011) yields up to 6 points, presentation on the essay topic (Apr. 28 - May 5, 2011) yields up to 4 points, weekly exercises may bring max 6 extra points.

Literature

Stuart Russell & Peter Norvig: Artificial Intelligence: A Modern Approach, Third ed., Pearson, 2010