Artificial Intelligence

Chapter four part 2

First-Order Logic
Topics we will cover
 Why FOL?
 Syntax and Semantics of FOL
 Using FOL
 Wumpus World in FOL
Introduction
 Propositional Logic: Represent a knowledge base (KB) using a
formal language.
 Has efficient linear complexity inference algorithms that
allow us to reason with the knowledge in the KB,
 It is limited in its expressiveness.
 First-Order Logic:
 An alternative (and more expressive) language for
representing KBs.
Good points () and bad points () of
propositional logic
 Propositional logic is declarative:
 The order of knowledge in KB not important.
 Propositional logic allows partial/disjunctive/negated information:
 (unlike most data structures and databases)
 Propositional logic is compositional:
 meaning of B1,1  P1,2 is derived from meaning of B1,1 and of P1,2

 Propositional logic has very limited expressive power,

 E.g., cannot say “pits cause breezes in adjacent squares”
 except by writing one sentence for each square
 First-Order Logic (FOL)
 Propositional logic assumes the world contains facts,
 First-order logic (like natural language) assumes the world contains,
 Objects : represent things in the real world.
 E.g: people, houses, numbers, colors …
 Relations : represent attributes of object or relationships between
objects.
 E.g: red, round, prime, brother of, bigger than, part of, comes
between, …
 Functions : take object(s) and return an object result.
 E.g: Length(), TeammateOf(), best friend(), Sqrt() …
Syntax of FOL: Basic elements
 Sentences in FOL represent these components of the world using
the following elements:
 constants: represent objects
 E.g: Haileselassie, 2, Blue, …
 predicates: represent relations
 E.g: Brother, >, …
 functions: represent functions
 E.g: Sqrt, LeftLegOf, …
 variables: by convention lower case letters
 E.g: x, y, a, b,...
 connectives: the same 5 connectives as are used
 i.e.: , , , , 

 equality: the two terms refer to the same object, i.e. =

 quantifiers: existential and universal, i.e.  ,
 Atomic Sentences
 Atomic sentence:
 The simplest type of sentence in FOL.
 An atomic sentence consists of either:
 Atomic sentence = predicate (term1,...,termN)
or term1 = term2
 where a term is either:
 Term = function (term1,...,termN)
or constant or variable
 E.g.1. TeammateOf(Kenenisa,HaileGebreselassie)
 The first example consists of a predicate, TeammateOf, with two constant
arguments, HaileGebreselassie and Kenenisa.
 E.g.2 Length(LeftLegOf(Kenenisa)) = Length(RightLegOf(Kenenisa))
 The second example contains two constants (both Kenenisa), combined
using three functions, Length, LeftLegOf and RightLegOf.
Complex Sentences
 Complex sentences are made from atomic sentences using
connectives:
 We can make more complex sentences by combining atomic
sentences using connectives,
S, S1  S2, S1  S2, S1  S2, S1  S2,
 E.g.
 Teammate(Kenenisa,HaileGebreselassie)
Teammate(HaileGebreselassie,Kenenisa)
 >(1,2)  ≤ (1,2)
 >(1,2)   >(1,2)
Interpretations
 Sentences are true with respect to a model and an
interpretation.
 Model contains objects (domain elements) and relations
among them.
 Interpretation specifies (real-world) referents for
constant symbols → objects
predicate symbols → relations
function symbols → functional relations

 An atomic sentence predicate(term1,...,termN) is true

if the objects referred to by term1,...,termN are in the relation

referred to by predicate.
First-Order Logic Quantifiers
 First-Order Logic Quantifiers
 It is FOL quantifiers that give the language its expressive
power.
 There are two types of quantifier:
 universal quantifiers ()
 existential quantifiers ()
Universal Quantification
 The symbol for universal quantification is .
 To interpret the universal quantification symbol in a FOL
sentence we can read it as “for all …”.
 <variables> <sentence>
 E.g. x At(CS , x)  Intelligent(x)
 Can be interpreted as stating that “for all x, At(CS, x)
implies Intelligent(x)”.
 In other words, if everyone is at CS then they are
intelligent.
 Definition:
 x P is true in a model m iff P is true with x being each
possible object in the model.
A common mistake to avoid
 Typically,  is the main connective with .
 Common mistake: using  as the main connective with :
 For Examples:
 x At(CS , x)  Intelligent(x)
 Means “Everyone is at CS and everyone is intelligent”
 x At(CS , x)  Intelligent(x)
 Means “Everyone at CS is intelligent”
Existential Quantification
 The symbol for existential quantification is .
 We can read this as “there exists …”.
 <variables> <sentence>
 E.g. x At (CS , x)  Intelligent(x)
 can be interpreted as stating that “there exists some x, such
that At(CS, x) and Intelligent(x) are both true.
 In other words, somebody at CS is intelligent.
 Definition:
 x P is true in a model m iff P is true with x being some possible
object in the model
Another common mistake to avoid
 Typically,  is the main connective with .
 Common mistake: using  as the main connective with :
 For Examples:
 x At(CS , x)  Intelligent(x)
 is true if there is anyone who is not at CS!
 x At (CS , x)  Intelligent(x)

 Means “Someone at CS is intelligent”

 Properties of Quantifiers
 x y is the same as y x
 x y is the same as y x
 x y is not the same as y x
 x y Loves(x,y)
 “There is a person who loves everyone in the world”
 y x Loves(x,y)
 “Everyone in the world is loved by at least one person”
 Quantifier duality: each can be expressed using the other.
x Likes(x,dorowet) x Likes(x,dorowet)
x Likes(x, porridge) x Likes(x, porridge)
 Equality in First-Order Logic
 The equality symbol in FOL does not mean that same as assignment
in most programming languages.
 A sentence involving the ‘=’ symbol in FOL is true if both of its
arguments refer to the same object.
 term1 = term2 is true under a given interpretation if and only if term1
and term2 refer to the same object.
 E.g., x,y Brother(x,Abebe)  Brother(y,Abebe)  (x=y)
 x and y do not refer to the same object. In other words, Abebe has two
(distinct) brothers.
 Without the use of the equality symbol, the above sentence would be true
if x and y had the same value (i.e. Abebe has one brother).
 Interacting with FOL KBs
 Suppose a wumpus-world agent is using a FOL KB and perceives a smell and
a breeze (but no glitter) at t=5:
Tell(KB,Percept([Smell,Breeze,None],5))
Ask(KB,a BestAction(a,5))
 Tell/Ask Operations in First-Order Logic
 The first operation tells the KB, the current percept at time 5.
 The second operation asks the KB what will be the best action to take at
time 5.
 i.e., does the KB entail some best action at t=5?
 Answer: Yes, {a/Shoot} ← substitution (binding list)
 Given a sentence S and a substitution σ,
Sσ denotes the result of plugging σ into S; e.g.,
S = Faster(x,y)
σ = {x/Kenenisa, y/HaileGebreselassie}
Sσ = Faster(Kenenisa, HaileGebreselassie)
 Ask(KB,S) returns some/all σ such that KB╞ Sσ
Knowledge base for the wumpus world (1)
 Perception
 t,s,b Percept([s,b,Glitter],t)  Glitter(t)

 Reflex
 t Glitter(t)  BestAction(Grab,t)
Knowledge base for the wumpus world (2)
 Adjacency
 x,y,a,b Adjacent([x,y],[a,b]) 
[a,b]  {[x+1,y], [x-1,y],[x,y+1],[x,y-1]}

 Squares are breezy near a pit:

 Diagnostic rule---infer cause from effect
s Breezy(s)   r Adjacent(r,s)  Pit(r)
 Causal rule---infer effect from cause
r Pit(r)  [s Adjacent(r,s)  Breezy(s) ]
Summary
 First-order logic:
 objects and relations are semantic primitives.
 syntax: constants, functions, predicates, equality,
quantifiers.
 FOL:
 Increased expressive power: sufficient to define wumpus
world.
 Propositional logic lacks expressive power …