Skip to content
Sahithyan's S3
1
Sahithyan's S3 — Artificial Intelligence

Propositonal Logic

Less ontological commitment. Weak epistemological commitment.

Sentences are formed using logical connectives such as not, and.

  • Atomic sentence: a single proposition (e.g., PP).
  • Literal: an atomic sentence or its negation.
  • Complex sentence: built from simpler ones with connectives.

Defines the truth of sentences under a model (assignment of true/false values).

Example: m1={P=true,Q=true,R=false}m1 = \set{P = \text{true}, Q = \text{true}, R = \text{false}}

Aka. CNF. A standard way of writing logical formulas so that they can be easily processed by reasoning algorithms like resolution.

A formula is in CNF iff:

  • it’s a conjunction (\land) and
  • each clause inside the conjunction is a disjunction (\lor) of literals
  • Eliminate biconditionals (    \iff)
    Replace: (A    B)(A    B)(B    A)(A \iff B) \to (A \implies B) \land (B \implies A)

  • Eliminate implications (    \implies)
    Replace: (A    B)¬AB(A \implies B) \to \lnot A \lor B

  • Move lnot inwards using De Morgan’s laws

    • ¬(AB)(¬A¬B)\lnot(A \land B) \to (\lnot A \lor \lnot B)
    • ¬(AB)(¬A¬B)\lnot(A \lor B) \to (\lnot A \land \lnot B)
    • ¬(¬A)A\lnot(\lnot A) \to A
  • Apply distributive law to get a conjunction of disjunctions
    (A(BC))(AB)(AC)(A \lor (B \land C)) \to (A \lor B) \land (A \lor C)

  • Simplify
    Remove duplicate literals or clauses if possible.