(L03 links: changed format)
(28 intermediate revisions by the same user not shown)
Riadok 5: Riadok 5:
 
__TOC__
 
__TOC__
  
The course objectives are to make the students familiar with basic principles of various computational methods of data processing that can commonly be called computational intelligence (CI). This includes mainly bottom-up approaches to solutions of (hard) problems based on various heuristics (soft computing), rather than exact approaches of traditional artificial intelligence based on logic (hard computing). Examples of CI are nature-inspired methods (artificial neural networks, evolutionary algorithms, fuzzy systems), as well as probabilistic methods and reinforcement learning. After the course the students will be able to conceptually understand the important terms and algorithms of CI, and choose appropriate method(s) for a given task. The theoretical introduction is combined with practical examples.  
+
The course objectives are to make the students familiar with basic principles of various computational methods of data processing that can commonly be called computational intelligence (CI). This includes mainly bottom-up approaches to solutions of (hard) problems based on various heuristics (soft computing), rather than exact approaches of traditional artificial intelligence based on logic (hard computing). Examples of CI are nature-inspired methods (artificial neural networks, evolutionary algorithms, fuzzy systems), as well as probabilistic methods and reinforcement learning. After the course the students will be able to conceptually understand the important terms and algorithms of CI, and choose appropriate method(s) for a given task. The theoretical lectures are combined with the seminar where the important concepts will be discussed and practical examples will be shown.
 
<!--[https://sluzby.fmph.uniba.sk/infolist/sk/2-AIN-132_15.html Informačný list predmetu]-->
 
<!--[https://sluzby.fmph.uniba.sk/infolist/sk/2-AIN-132_15.html Informačný list predmetu]-->
  
<!--
 
 
== News ==
 
== News ==
<!--
 
'''Skúška:''' v AISe sú vypísané termíny (7., 9., 15., 20., 26.6), so začiatkom písomnej časti vždy o 8:30 v I-8. Môžete sa prihlasovať, maximálne 8 ľudí na termín. Odhlasovanie (prihlasovanie) je možné do 8:30 (18:30) v deň pred skúškou. Ťaháte si pseudonáhodne 3 otázky zo [http://dai.fmph.uniba.sk/courses/NN/ns-otazky.pdf zoznamu otázok].
 
  
;2. 9. 2017
+
'''Exam:''' The list of questions is [http://dai.fmph.uniba.sk/courses/ICI/ci-exam-questions.pdf here]. You will choose three questions (pseudo)randomly.
: We start on Monday, 25th September.
+
 
 +
<!--
 +
;11. 9. 2018
 +
: We start on Monday, 24th September.
 
<!-- [[#Archív noviniek|Archív noviniek…]] -->
 
<!-- [[#Archív noviniek|Archív noviniek…]] -->
  
Riadok 29: Riadok 29:
 
|Lecture
 
|Lecture
 
|Monday
 
|Monday
|9:00
+
|9:00 - 10:30
 
|I-9
 
|I-9
 
|[[Igor Farkas|Igor Farkaš]]
 
|[[Igor Farkas|Igor Farkaš]]
 
|-
 
|-
|Practical demonstrations
+
|Seminar
|Monday
+
|Wednesday
|10:30
+
|14:00 - 15:30
 
|I-9
 
|I-9
|[[Tomas Kuzma|Tomáš Kuzma]]
+
|[[Igor Farkas|Igor Farkaš]] & [https://www.sav.sk/index.php?lang=en&doc=user-org-user&user_no=11408 Xenia Daniela Poslon]  
 
|}
 
|}
  
Riadok 48: Riadok 48:
 
!References
 
!References
 
|-
 
|-
|25.09.
+
|24.09.
 
|What is computational intelligence, basic concepts, relation to artificial intelligence.  [http://dai.fmph.uniba.sk/courses/ICI/References/ci-def.4x.pdf slides]
 
|What is computational intelligence, basic concepts, relation to artificial intelligence.  [http://dai.fmph.uniba.sk/courses/ICI/References/ci-def.4x.pdf slides]
|Craenen & Eiben (2003); [https://en.wikipedia.org/wiki/Computational_intelligence wikipedia]; R&N (2010), chap.1
+
|Craenen & Eiben (2003); [https://en.wikipedia.org/wiki/Computational_intelligence wikipedia]; R&N (2010), chap.1; Sloman (2002)
 
|-
 
|-
|02.10.
+
|01.10.
|Taxonomy of artificial agents, nature of environments.
+
|Taxonomy of artificial agents, nature of environments. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-agents.4x.pdf slides]
 
|R&N (2010), chap.2
 
|R&N (2010), chap.2
 
|-
 
|-
|09.10.
+
|08.10.
|Inductive learning via observations, decision trees. Model selection.
+
|Inductive learning via observations, decision trees. Model selection. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-learning.4x.pdf slides]
 
|R&N (2010), ch.18.1-3,18.6; Marsland (2009), ch.6.1-2, [http://www.r2d3.us/visual-intro-to-machine-learning-part-1/ visualization], [http://fiddle.jshell.net/92Jxj/show/light/ interactive demo]
 
|R&N (2010), ch.18.1-3,18.6; Marsland (2009), ch.6.1-2, [http://www.r2d3.us/visual-intro-to-machine-learning-part-1/ visualization], [http://fiddle.jshell.net/92Jxj/show/light/ interactive demo]
 
|-
 
|-
|16.10.
+
|15.10. / 18.10.
|Supervised learning in feedforward neural networks (perceptrons), pattern classification, function approximation.
+
|Supervised learning in feedforward neural networks (perceptrons), pattern classification, function approximation. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-fwdnn.4x.pdf slides]
 
|R&N (2010), ch.18.2; Marsland (2009), ch.2-3, Engelbrecht (2007), ch.2-3
 
|R&N (2010), ch.18.2; Marsland (2009), ch.2-3, Engelbrecht (2007), ch.2-3
 
|-
 
|-
|23.10.
+
|24.10.
|Unsupervised (self-organizing) neural networks: feature extraction, data visualization.
+
|Unsupervised (self-organizing) neural networks: feature extraction, data visualization. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-unsup.4x.pdf slides]
 
|Marsland (2009), ch.9-10, Engelbrecht (2007), ch.4
 
|Marsland (2009), ch.9-10, Engelbrecht (2007), ch.4
 
|-
 
|-
|30.10.
+
|29.10.
|mid-term exam
+
|No class
|your mind :-)
+
|fall break (the whole week)
 +
|-
 +
|05.11.
 +
|Interim summary (review)
 +
|Wednesday: mid-term test
 
|-
 
|-
|06.11.
+
|12.11.
|Statistical learning, probabilistic models.
+
|Statistical learning, probabilistic models. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-prob.4x.pdf slides]
 
|R&N (2010), ch.13,20.1-2; Marsland (2009), ch.8.1-2
 
|R&N (2010), ch.13,20.1-2; Marsland (2009), ch.8.1-2
 
|-
 
|-
|13.11.
+
|19.11.
|Reinforcement learning I: basic principles and learning methods (TD-learning). Prediction problem.
+
|Reinforcement learning I: basic principles and learning methods (TD-learning). Prediction problem. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-rl.4x.pdf slides]
|R&N (2010), ch.21.1-2.
+
|R&N (2010), ch.21.1-2.  
 
|-
 
|-
|20.11.
+
|26.11.
 
|Reinforcement learning II (Q, SARSA), actor-critic, control problem, RL for continuous domains.
 
|Reinforcement learning II (Q, SARSA), actor-critic, control problem, RL for continuous domains.
|R&N (2010), ch.21.3-5.
+
|R&N (2010), ch.21.3-5; Woergoetter & Porr (2008).
 
|-
 
|-
|27.11.
+
|03.11.
|Evolutionary computation: basic concepts, genetic algorithms.
+
|Evolutionary computation: basic concepts, genetic algorithms. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-evol.4x.pdf slides]
 
|Engelbrecht (2007), ch.8
 
|Engelbrecht (2007), ch.8
 
|-
 
|-
|04.12.
+
|10.12.
|Fuzzy systems, fuzzy logic and reasoning.
+
|Fuzzy systems, fuzzy logic and reasoning. [http://dai.fmph.uniba.sk/courses/ICI/References/ci-fuzzy.4x.pdf slides]
|Engelbrecht (2007), ch.20-21, Scholarpedia: Zadeh (2007)  
+
|Engelbrecht (2007), ch.20-21; Zadeh (2007)  
 
|}
 
|}
  
Riadok 96: Riadok 100:
  
 
* Craenen B., Eiben A. (2003): [http://dai.fmph.uniba.sk/courses/ICI/craenen.ci.enc03.pdf Computational Intelligence]. In: Encyclopedia of Life Support Sciences, EOLSS Publishers Co.
 
* Craenen B., Eiben A. (2003): [http://dai.fmph.uniba.sk/courses/ICI/craenen.ci.enc03.pdf Computational Intelligence]. In: Encyclopedia of Life Support Sciences, EOLSS Publishers Co.
* Engelbrecht A. (2007). Computational Intelligence: An Introduction (2nd ed.), John Willey & Sons. Available in faculty library.
+
* Engelbrecht A. (2007). [http://dai.fmph.uniba.sk/courses/ICI/engelbrecht.comp-intel-intro.07.pdf Computational Intelligence: An Introduction] (2nd ed.), John Willey & Sons.
* Russell S., Norwig P. (2010). [http://dai.fmph.uniba.sk/courses/ICI/russell-norvig.AI-modern-approach.3rd-ed.2010.pdf Artificial Intelligence: A Modern Approach], (3rd ed.). Available in the faculty library.
+
* Russell S., Norwig P. (2010). [http://dai.fmph.uniba.sk/courses/ICI/References/russell-norvig.AI-modern-approach.3ed.2010.pdf Artificial Intelligence: A Modern Approach], (3rd ed.), Prentice Hall. Available in the faculty library.
* Marsland S. (2009). [http://www-ist.massey.ac.nz/smarsland/MLBook.html Machine Learning: An Algorithmic Perspective], CRC Press. Available in the faculty library.
+
* Marsland S. (2015). [http://dai.fmph.uniba.sk/courses/ICI/References/marsland.machine-learning.2ed.2015.pdf  Machine Learning: An Algorithmic Perspective], (2nd ed.), CRC Press.
* Zadeh L. (2007). [http://dx.doi.org/10.4249/scholarpedia.1766 Fuzzy logic], Scholarpedia, 3(3):1766.  
+
* Sloman A. (2002). [http://www.cs.bham.ac.uk/research/projects/cogaff/sloman.turing.irrelevant.pdf The Irrelevance of Turing Machines to AI]. In Scheutz M. (ed.): Computationalism: New Directions, MIT Press, Cambridge, MA, pp. 87–127.
 +
* Woergoetter F., Porr B. (2008). [http://www.scholarpedia.org/article/Reinforcement_learning Reinforcement learning], Scholarpedia, 3(3):1448.
 +
* Zadeh L. (2007). [http://www.scholarpedia.org/article/Fuzzy_logic Fuzzy logic], Scholarpedia, 3(3):1766.  
  
 
== Course grading ==
 
== Course grading ==
  
* Active participation during the semester (max. 14 points).
+
* Active participation during the seminar/exercise: 5 for lectures, 10 for exercises (max. 15 points). It includes optional final project (implementation of a small neural network and writing a short report ("optional" means that the student can still get maximum points if s/he has performed very actively).
* Written mid-term test (max. 12 points).
+
* Written mid-term test (max. 15 points).
* Final written-oral exam (max. 24 points, 3 questions).  
+
* Final written-oral exam (max. 20 points, 3 questions). Minimum of 5 points required.
 
* <b>Overall grading:</b> A (50-46), B (45-41), C (40-36), D (35-31), E (30-26), Fx (25-0).
 
* <b>Overall grading:</b> A (50-46), B (45-41), C (40-36), D (35-31), E (30-26), Fx (25-0).

Verzia zo dňa a času 09:18, 2. január 2019

Introduction to Computational Intelligence 2-IKV-115

The course objectives are to make the students familiar with basic principles of various computational methods of data processing that can commonly be called computational intelligence (CI). This includes mainly bottom-up approaches to solutions of (hard) problems based on various heuristics (soft computing), rather than exact approaches of traditional artificial intelligence based on logic (hard computing). Examples of CI are nature-inspired methods (artificial neural networks, evolutionary algorithms, fuzzy systems), as well as probabilistic methods and reinforcement learning. After the course the students will be able to conceptually understand the important terms and algorithms of CI, and choose appropriate method(s) for a given task. The theoretical lectures are combined with the seminar where the important concepts will be discussed and practical examples will be shown.

News

Exam: The list of questions is here. You will choose three questions (pseudo)randomly.


Course schedule

Type Day Time Room Lecturer
Lecture Monday 9:00 - 10:30 I-9 Igor Farkaš
Seminar Wednesday 14:00 - 15:30 I-9 Igor Farkaš & Xenia Daniela Poslon


Syllabus

Date Topic References
24.09. What is computational intelligence, basic concepts, relation to artificial intelligence. slides Craenen & Eiben (2003); wikipedia; R&N (2010), chap.1; Sloman (2002)
01.10. Taxonomy of artificial agents, nature of environments. slides R&N (2010), chap.2
08.10. Inductive learning via observations, decision trees. Model selection. slides R&N (2010), ch.18.1-3,18.6; Marsland (2009), ch.6.1-2, visualization, interactive demo
15.10. / 18.10. Supervised learning in feedforward neural networks (perceptrons), pattern classification, function approximation. slides R&N (2010), ch.18.2; Marsland (2009), ch.2-3, Engelbrecht (2007), ch.2-3
24.10. Unsupervised (self-organizing) neural networks: feature extraction, data visualization. slides Marsland (2009), ch.9-10, Engelbrecht (2007), ch.4
29.10. No class fall break (the whole week)
05.11. Interim summary (review) Wednesday: mid-term test
12.11. Statistical learning, probabilistic models. slides R&N (2010), ch.13,20.1-2; Marsland (2009), ch.8.1-2
19.11. Reinforcement learning I: basic principles and learning methods (TD-learning). Prediction problem. slides R&N (2010), ch.21.1-2.
26.11. Reinforcement learning II (Q, SARSA), actor-critic, control problem, RL for continuous domains. R&N (2010), ch.21.3-5; Woergoetter & Porr (2008).
03.11. Evolutionary computation: basic concepts, genetic algorithms. slides Engelbrecht (2007), ch.8
10.12. Fuzzy systems, fuzzy logic and reasoning. slides Engelbrecht (2007), ch.20-21; Zadeh (2007)

References

Course grading

  • Active participation during the seminar/exercise: 5 for lectures, 10 for exercises (max. 15 points). It includes optional final project (implementation of a small neural network and writing a short report ("optional" means that the student can still get maximum points if s/he has performed very actively).
  • Written mid-term test (max. 15 points).
  • Final written-oral exam (max. 20 points, 3 questions). Minimum of 5 points required.
  • Overall grading: A (50-46), B (45-41), C (40-36), D (35-31), E (30-26), Fx (25-0).