Skip to content

Latest commit

 

History

History
52 lines (37 loc) · 1.65 KB

Chapter6-logic.md

File metadata and controls

52 lines (37 loc) · 1.65 KB

Chapter 6, Logical agents

Knowledge Representation

What is logic? A way to make destinctions between good and bad arguments and the tools for making that distinction.

word-list

  • Semantics - study of meaning, focus on associating elements of formal language with elements of a domain of discourse.
    • well-formed formulas = sentences - A word that is part of a formal language
    • proposition - is a sentence expressing if something is true or false
    • Interpretation associates a propositon with an atom
    • atom - simplest well-formed formulas of logic
  • Domain of discourse - what sentences can our set of entities be? Ex natural numbers, car colours, natural numbers are all exambles of domains of discourse.

===

Diffrent types of logic

Propositional logic:

  • functions: "and" "or" "implies" "not" ""
    • and - w ^ u
    • or - W v u
    • implication - !W v u
    • not - !w
  • Domain of discourse - propositions about the world
  • atoms - variables

Logic list

& = set of sentences, W = possible worlds, RW = real world

  • Syntax
    • Inference - & |-- RW
    • Entailment - & |= W

===

Theorem proving, inference and proofs

  • Logical equivalence - two sentences are true in the same sets of models a<=>b
  • Validity - A sentence is valid if it's true in all models. a v !a
  • satisfiability - a sentence is satisfiable if it's true in some model

CNF

How to convert a sentence to CNF form

  1. eliminate a <=> b by replacing with (a=>b)^(b=>a)
  2. eliminate a=>b by replacing with !a v b
  3. CNF requires ! to appear only in litterals so !(!a)=a, !(avb)=(!a)v(!b)
  4. We want to distribute v over ^ whenever its possible (a v (b^c)=(avb)^(avc)