This is last year G53KRR web page!

This year G53KRR page is on moodle. The material is similar with the addition of reasoning about actions and planning, and added coursework. This year G53KRR is 20 credits.

Knowledge Representation and Reasoning 2017-2018

Feedback on G53KRR exam

Recent paper on importance of KRR in AI curriculum

The module describes how knowledge can be represented symbolically and manipulated in an automated way by reasoning programs. It will be based on the following textbook: Ronald Brachman and Hector Levesque. Knowledge Representation and Reasoning. Elsevier, 2004. There are several copies in the Jubilee Library, location Q387 BRA. And there is a link to eBook edition from the library catalogue.

Hector Levesque's on-line lecture slides based on the book
local copy

Other learning resources

• Michael Huth and Mark Ryan, Logic in Computer Science Chapters 1 and 2 cover propositional and predicate logic. There are exercises on the web page. The book is in the library: DLRC, Jubilee Campus QA76.9.L63 HUT. There is an electronic copy kindly provided by Mark Ryan on G53KRR moodle page. Please only use this electronic copy for your own study of logic and do not share copies on the internet.
• Stuart J. Russell and Peter Norvig, Artificial intelligence : a modern approach. In any of the editions, there is a chapter or part on Knowledge and Reasoning (you need first order logic). Lots of copies in the library: DLRC, Jubilee Campus Q335 RUS, Business Library Q335.R8.
• D. Partridge, K.M. Hussain. Knowledge-based information systems. London : McGraw-Hill, 1995, Ch.6,7. (development cycle, decision tables)
• E. Rich, K. Knight. Artificial Intelligence. McGraw Hill, 1991. Ch. 20.4 (Knowledge Acquisition).
• Enrico Franconi's Description Logic course (Modules 2 and 3)
• Peter F. Patel-Schneider, Building the Semantic Web Tower from RDF Straw, In Proceedings IJCAI 2005

Office hour (for questions): Thursdays 16:00-17:00 in Computer Science C57

Lectures

• Lecture 1: module information, introduction
• Lecture 2: syntax of first-order logic. See Levesque's slides 18-27, textbook chapter 2.1 - 2.3.2.
• Lecture 3: semantics of first-order logic, logical entailment. See Levesque's slides 24-34, textbook the rest of chapter 2.
• Lecture 4: semantics of first-order logic continued. Same slides as before, especially the definition of being true in an interpretation (slides 27-29).
  Barber exercise
• Lecture 5: entailment, satisfiability, valilidity. Expressing knowledge in first-order logic. Levesque's slides 36-45, texbook chapter 2, 3.
• No lecture on 19 October.
  Exercises on first order logic (please do the exercises first and look at the answers later. If you only do one, do Alpine Club.):
  • Barber exercise
  • Answer to the barber exercise
  • Informal exercise on syntax and semantics of first-order logic
  • Answer to the exercise on syntax and semantics of first-order logic
  • Knowledge representation exercise: Alpine Club exercise
  • Answer to Alpine Club exercise
  • Question 1 from any exam paper from previous years
• Lecture 6: answer to Alpine Club exercise. Early module feedback.
• Lecture 7: propositional resolution. Levesque's slides 46-55, textbook chapter 4.1.

  Handout with definitions and rules

  Exercise on propositional reduction to CNF
  Answer to the exercise
  Exercise on propositional resolution
  Answer to the exercise

• Lecture 8: First-order resolution. Chapter 4.2 in the textbook. Slides 56-66.

  Handout with definitions and rules
  Exercise on first-order resolution. (For now, only do reduction to clausal form)
  Answer to the exercise on first-order resolution.

• Lecture 9: First order resolution continued. Slides 56-70, textbook 4.2, 4.3 (apart from Herbrand theorem and complexity)
  Exercise on unification: do question 3 of 2013-14 exam
  Answer to the exercise on unification: see answer to question 3 of 2013-14 exam
• Lecture 10: Answer extraction. Complexity of resolution. Slides 62-63, 66-67, 72-78.
  Exercise on answer extraction
  Answer to the exercise
• Lecture 11: Horn clauses. SLD resolution. Backward chaining. Chapter 5, slides 79-90. Exercise: do question 4 in 2010-2011 and 2011-2012 exam. (Exams and answers are here )
• Lecture 12: finish backward chaining; Prolog. Reading: Chapter 6, slides 91-100. You can try out Prolog here . SWI Prolog is still probably installed in the lab.
• Lecture 13: Forward chaining. Rules in production systems. Chapter 7, slides 89, 103-118.
  Handout about forward chaining and exercise
  Answer to the exercise
• Lecture 14: Life cycle of a knowledge based system. Slides For an exercise, you can do exam 2007 (5b), 2008 (4d), 2009 (6b), 2011 (5a).
• Lecture 15: description logic. Handout . See Other learning resources above for more reading. Example ontologies:
  DBPedia , Google's Knowledge Graph , Snomed CT
  Exercise
  Answer
• Exercises from previous exams, annotated for ALC. and answers
• 27 November: exercises on description logic, tutorial given by Rustam Galimullin.
• Lecture 16: non-monotonic reasoning. Chapter 11, slides 178-195.
• Lecture 17: Non-monotonic reasoning continued: defaults.
  Summary of lectures 16 and 17 (Updated 7/12/17)
  For exercises, do exam questions from previous years:
  • 2013: question 5 (on defaults, can only do it after lecture 17),
  • 2012: question 5 (CWA and circumscription),
  • 2010: question 5 (CWA and circumscription),
  • 2009: question 5 (CWA and circumscription),
  • 2008: question 6 (circumscription and defaults).
  Exam papers and answers are here.
• Lecture 18: Bayesian networks. Chapter 12, slides 202-216.
  Summary of Bayesian networks and an exercise
  Solution to the exercise
• 11 December: last lecture. SEM feedback and exam revision advice.
• No lecture on 14 December. Revision lecture on the 12th of January also cancelled, sorry.