Summary: Knowledge Representation

Read the summary and the most important questions on Knowledge Representation

• Lecture 1

  This is a preview. There are 6 more flashcards available for chapter 07/09/2016
  Show more cards here

• What are definitions of intelligence?

  
  Carry out complex reasoning
  Draw...
  slides

• What are the two types of AI?

  Statistical and symbolic AI.

• What are applications for statistical AI?

  Pattern recognition (images, sound, shapes).
  Motor skills (robots).
  Speech generation (sound).
  Search engines.

• What are applications for symbolic AI?

  Planning (autonomous space missions).
  Reasoning (diagnosis, design, decision support).
  Language generation (conversations).
  Search engines.

• Lecture 2

• To what form can everything be reduced?

  Clausal normal form (CNF), a conjunction of disjunction.
  1. Remove implications
  2. Move negations inward
  3. Move conjunction outward (result: conjunction of clauses)
  4. Split up conjunctive clauses.

• Propositional logic is a weak language. Why?

  Propositional letters only describe complete "states" of the world.
  We cannot talk about "individuals".
  Can't directly talk about properties of individuals or relations between individuals.
  Generalisations, patterns, regularities can't easily be represented. (all triangles have   3 sides)
  First-Order Logic (FOL or FOPC) is expressive enough to concisely represent this kind of information.
  
  
  Static finite universe.

• Statistiability Testing (SAT)

  Given: a set of clauses C
  Decide: Is C satisfiable (does there exist a model M for C)
  Construct at least one such M
  
  Answer = M
  (assignment of {1,0} to all variables which makes C true)

• What is the DP algorithm?

  It needs a:
  Unit clause: a clause containing a single literal; the only way to satisfy such a clause is by assigning its literal the true value
  SOMETHING ELSE Pure literal
  
  
  Conventions:
  If omgekeerde v = 0 T (empty??)
  If V = 0 F
  
  Example:
  
  Werk uit uit schrift

• Lecture 3

  This is a preview. There are 19 more flashcards available for chapter 13/09/2016
  Show more cards here

• What is CN(x)?

  The number of occurrences of x' in unresolved clauses.

• What are Jeroslow-Wang heuristics?

  For a given literal I compute:
  function""
  One sided Jeros