Artificial Intelligence
Code: 104398 ECTS Credits: 6| Degree | Type | Year | Semester |
|---|---|---|---|
| 2503740 Computational Mathematics and Data Analytics | OB | 2 | 2 |
Contact
- Name:
- Maria Vanrell Martorell
- Email:
- Maria.Vanrell@uab.cat
Use of Languages
- Principal working language:
- catalan (cat)
- Some groups entirely in English:
- No
- Some groups entirely in Catalan:
- Yes
- Some groups entirely in Spanish:
- No
Prerequisites
Given the introductory nature of this subject, we assume the student does not have any previous knowledge about this topic. Is the aim of the subject to give to the students the means to acquire the knowledge contents described in the contents section of this guide.
About other skills we expect from the students:
- Having coursed a technological or scientific background in the secondary school.
- Having passed the programming subjects of the 1st and 2nd year of this degree.
- Knowing basic notions of python programming
- Having skills at a user level of one of the following platforms (Windows, Mac or Linux)
Objectives and Contextualisation
This is a subject devoted to introducing the Student to the artificial intelligence (AI) field. Two main objectives are pursuit:
a) let the students to learn how resolution of AI problems is performed with their own specificities in representation, evaluation and solving methodologies, and
b) let the students to know a wide range of techniques and basic algorithms that allow to solve the proposed problems and improving their programming skills.
More specifically, these two aims are pursuing:
- Giving a basic historical introduction and aims of the AI field
- Facing the student with the problem of selecting a good knowledge representation as the basic step to solve an AI problem
- Familiarize the student in 4 different knowledge representations and their corresponding algorithms.
- Giving to the student the ability to design solutions to contextualized problems
- Giving to the student the ability to present and justify the adequacy of the designed solutions.
Competences
- Make effective use of bibliographical resources and electronic resources to obtain information.
- Solve problems related to the analysis of large volumes of data through the design of intelligent systems and computational learning.
- Students must be capable of applying their knowledge to their work or vocation in a professional way and they should have building arguments and problem resolution skills within their area of study.
- Students must be capable of communicating information, ideas, problems and solutions to both specialised and non-specialised audiences.
- Students must have and understand knowledge of an area of study built on the basis of general secondary education, and while it relies on some advanced textbooks it also includes some aspects coming from the forefront of its field of study.
- Using criteria of quality, critically evaluate the work carried out.
- Work cooperatively in a multidisciplinary context assuming and respecting the role of the different members of the team.
Learning Outcomes
- Define and develop optimal search schemes for distinct problems based on the representation and classification of knowledge.
- Identify and define computational solutions in multiple domains for decision making based on exploring alternatives, uncertain reasoning and task planning.
- Know and understand techniques for representing human knowledge.
- Make effective use of bibliographical resources and electronic resources to obtain information.
- Students must be capable of applying their knowledge to their work or vocation in a professional way and they should have building arguments and problem resolution skills within their area of study.
- Students must be capable of communicating information, ideas, problems and solutions to both specialised and non-specialised audiences.
- Students must have and understand knowledge of an area of study built on the basis of general secondary education, and while it relies on some advanced textbooks it also includes some aspects coming from the forefront of its field of study.
- Using criteria of quality, critically evaluate the work carried out.
- Work cooperatively in a multidisciplinary context, taking on and respecting the role of the distinct members in the team.
Content
1. Introduction to AI. Aim and brief history of the field. Definition of rational agent, methodologies for solving AI problems and knowledge representation.
2. Problem solving by Searching on alternatives
2.1. Informed Search. Basic definitions on search and review of non-informed search algorithms. Analysis of search algorithms: Completeness, Optimality and Complexity. Heuristic concept and examples. Informed Search: basic and optimal. A* and its properties. Effective branching factor and heuristic properties.
2.2. Local Search. Basic definitions, pros and cons. Understanding local search as exploring the heuristic function landscape. Local search with known goal: Hill-Climbing algorithm. Problems of local search: local maxima, plateaus and ridges. Local search with unknown goal: Steepest Ascent, Steepest Ascent with local maxima control, Simulated Annealing.
2.3. Adversarial Search. (a) Search based on Minimax. Basic definitions. Minimax algorithm and alpha-beta pruning. Complexity analysis. Minimax variations: Progressive deepening and Singular extensions. Examples of heuristics functions. (b) Search based on random simulations. Adversarial Search with Random Simulations. Basic definitions. Monte-Carlo Tree Search algorithm. Examples.
3. Solving pattern recognition problems.
3.1. Statistical approaches. Feature spaces as a representation for case-based reasoning. Basic definitions on feature spaces. Feature selection and dimension reduction. Classification based on supervised learning: decision functions, assumptions about the data distribution, local search to find decision function. Classification based on unsupervised learning: K-means algorithm, searching the best k with Fisher discriminant. Analysis of the algorithms.
3.2 Structural approaches. Graphs and semantic networks. Basic definitions and representation with adjacency matrices. The graph matching problem,basic algorithms, improvements (AC4) and complexity. Inexact graph matching: similarity measures, edition distance. Study case: String-matching.
4. Solving problems of logic reasoning
4.1. Logic and inference mechanisms. Knowledge representation: propositional logic and predicate logic. Review of basic algorithms: natural deduction, resolution mechanism, unification, clausal formconversion. Basic definitions and algorithms on rule-based Systems: rule base, working memory, rule chaining mechanisms, conflict resolution strategies.
4.2. Reasoning with Uncertainty. Representation of uncertaingy with fuzzy sets. Modus ponns with fuzzy sets and forward chaining.
Methodology
Artificial Intelligence is defined by the type of problems is trying to solve, thus in this course, the type of problem is organizing the course content. We will work in three different types of sessions:
Theoretical class: these are sessions where the lecturer directly explains the theoretical content of the course that will also be given in parallel thorugh videos. In these sessions explanation and evaluation of the learning process will be evaluated through quizes and interactive activities.
Problem-based class: these sessions try to facilitate interaction between students through the process of solving a problem that puts in practice the theoretical contents. In these sessions we pursuit to reinforce the understanding of the topics presented in theoretical sessions by posing practical cases that require the design of a solution using the methods presented in theory. It is impossible to follow these sessions without following theoretical sessions, since they are strongly linked.
Practical sessions: these sessions are organized to introduce and do the follow-up of the development of a the team-based project performed by the students. Different kind of activities are done in these sessions: (a) sessions for team work but tutored by teaching assistants, (b) sessions to individually evaluate through quizzes to the students on site, (c) sessions to present the results, where all the team members must explain and defend the results of the developed project.
Transversal skills we work on are:
- Acquiring thinking skills, specifically developing systemic thinking.
This skill is acquired from the theoretical contents, since all the contents are based on building intelligent system with interacting modules, as well as from the projects the students work on, since they work on teams that generate a lot of dependencies in their tasks. This skill is evaluatedin the theoretical exams, in the problem task delivered and at all the levels of the evaluation of the project.
- Acquiring personal work skills, specifically autonomous working.
This skill is acquired from the study of the theoretical contents, the individual delivery of the problem assignments and by the student participation in the project. In the three cases the individual work is evaluated, in theoretical contents through the exam, in problems through the evaluation of the deliverables and in the project through the individual quizzes, the participation in the final presentation and the intra-evaluation performed by the team members within each group.
Activities
| Title | Hours | ECTS | Learning Outcomes |
|---|---|---|---|
| Type: Directed | |||
| Theoretical lectures | 28 | 1.12 | 3, 1, 2, 7 |
| Type: Supervised | |||
| Practical sessions | 14 | 0.56 | 8, 6, 5, 9, 4 |
| Problem-based sessions | 12 | 0.48 | 3, 1, 2, 5 |
| Type: Autonomous | |||
| Individual study | 10 | 0.4 | 3, 1, 2, 7, 5, 4 |
| Project-based practical work | 22 | 0.88 | 8, 3, 1, 2, 6, 5, 9, 4 |
Assessment
To evaluate the learning level of the students we establish a formula that combines acquisition of knowledge, abilities in problem solving and the skills to work in a team as well as to justify the obtained results in a project.
Final grade is computed as a weighted sum of the marks obtained in the different performed activities:
Final Score = 0.5 * Theory Score + 0.1 * Problem Score + 0.4 * Project Score
This formula will be computed only in the case that Theory Score is greater or equal to 5, and the Project Score is greater or equal to 6. No constrain is applied on the Problem Score. If the Final Score is >=5, but the minimun thresholds are not achieved in on the the Scores, then the Score in the transcript will be 4.5.
The Theory Score evaluates the individual achievements of the student in the theoretical contents, this is performed in a continuous form during the course through several exams.
These exams pursuit to individually evaluate the student abilities in solving problems using the theoretical contents explained in the course, as well as, the level of abstraction achieved byt the student in the understanding of these techniques.
Retake Exam. In case the Theory Score does not achieve the 5 to pass, the student can perform a retak exam.
The Problem Score pursuits to provoke the student to work on the theoretical content as they are given during the course, specific problems are posed to make the student have to apply the theoretical contents just after they have been explained in the lectures. The solved problems are weekly delivered.
Problem Score = takes into account the number of deliverables and their correctness
This score is computed in the final mark if the % of deliverables is superior to 70% of the total.
The Project Score has an essential weight in the final mark, it pursuits the student to program and explore the studied techniques within the frame of the global goal of a contextualized project. Additionally, the student has to demonstrate its skills in doing all this work in a team and defending the final results. Two projects are performed, and the score is
Project Score = 0,5 * Projecte 1 Score + 0,5 * Project 2 Score
To compute this Score every one of the projects has to get a mark greater or equal to 6. The mark for each one of the projects is computed from a compilation of marks:
Project Score = 0,6 * Individual Mark + 0,4 * Group Mark
Group Mark has to be greater or equal to 5.
Individual Mark = 0,4 * Quiz_Mark + 0,3 * Individual_Presentation + 0,3 * Participation_in_groupl
Group Mark= 0,6 * Code_Mark + 0,3* Report_Mark + 0,1 * Group_Presentation
Quiz_Mark and Code_Mark have to be greater or equal to 5.
Retake Sessions. In case the Project score does not arrive to the adequate level to pass, the students will have two recovery options:
The option of delivering and defense again their projects in a specific retake session. In case the student requires to go to this retake session, then the maximal Group_Mark that can be obtained is 7.
The option of doing the quiz in a retake session. In case the student requires to go to this retake session, then the maximal Quiz_Mark that can be obtained is 7.
Important notes:
In case the subject is not passed due to one of the evaluation activities does not arrive to the minimum required score, the final score in the transcript will be the minimum between 4.5 and the final mark obtained if the threshold is not considered, with the exception that the numeric score in the transcript will be in between 3.0 and the final mark obtained if the threshold is not considered, for the case the student have performed any irregular act in an evaluation activity, such as those explained below.
Grade will be “Non-Graded” (“No Avaluable”) in the case the student did not participate in any of the evaluation activities.
Grade will be “Honors” (“Matrícula d’Honor”) in the case the rank of the grade is less than the maximum number of honors can be given in a course, and the value of the grade is over a threshold that will be stablished byt the teacher.
The evaluation and delivery dates will be published at caronte.uab.cat and might be shifted if there is any change in the planning due to any unexpected event. Students will beinformed about any change through caronte.uab.cat that will be the usual communication mean between students and teachers.
For the case of students retaking the course, no recognition with grades of the previous year will be considered.
Notwithstanding other disciplinary measures deemed appropriate, and in accordance with the academic regulations in force, assessment activities will receive a zero whenever a student commits academic irregularities that may alter such assessment. Assessment activities graded in this way and by this procedure will not be re-assessable. If passing the assessment activity or activities in question is required to pass the subject, the awarding of a zero for disciplinary measures will also entail a direct fail for the subject, with no opportunity to re-assess this in the same academic year. Irregularities contemplated in this procedure include, among others:
- the total or partial copying of a practical exercise, report, or any other evaluation activity;
- allowing others to copy;
- presenting group work that has not been done entirely by the members of the group;
- presenting any materials prepared by a third party as one’s own work, even if these materials are translations or adaptations, including work that is not original or exclusively that of the student;
- having communication devices (such as mobile phones, smart watches, etc.) accessible during theoretical-practical assessment tests (individual exams).
An overall grade of 5 or higher is required to pass the subject. A "non-assessable" grade cannot be assigned to students who have participated in any of the individual partial tests or the final exam.
No special treatment will be given to students who have completed the course in previous academic years, except that the seminar grade previously obtained can be assigned to this course gradebook.
The grade in the Transcript of Records (ToR) will be the lowest value between 3.0 and the weighted average grade, in the eventof irregularities having been committed for any assessment activity (and therefore re-assessment will not be possible).
Assessment Activities
| Title | Weighting | Hours | ECTS | Learning Outcomes |
|---|---|---|---|---|
| Individual Exam | 0.5 | 4 | 0.16 | 3, 1, 2, 7, 6, 5 |
| Project Defence (Quiz+Report+Code+Intra_group+Presentation) | 0.12 | 60 | 2.4 | 8, 1, 6, 5, 9, 4 |
| Solved Exercise Delivery | 0.1 | 0 | 0 | 3, 1, 2, 7, 6, 5 |
Bibliography
- S. Russell i P. Norvig, Artificial Intelligences - A modern approach. Prentice Hall, 2003, http://aima.cs.berkeley.edu/
- Tveter, Donald R., (1998), The Pattern Recognition basis of Artificial Intelligence. IEEE Computer Society.