12.2.
1 Resolution Principle(1)
• Resolution refutation proves a theorem by
negating the statement to be proved and adding
this negated goal to the set of axioms that are
known to be true.
• Use the resolution rule of inference to show that
this leads to a contradiction.
• Once the theorem prover shows that the negated
goal is inconsistent with the given set of axioms, it
follows that the original goal must be consistent.
Resolution Theorem Proving 2
12.2.1 Resolution Principle(2)
• Steps for resolution refutation proofs
• Put the premises or axioms into clause form(12.2.2).
• Add the negation of what is to be proved, in clause form, to
the set of axioms.
• Resolve these clauses together, producing new clauses that
logically follow from them(12.2.3).
• Produce a contradiction by generating the empty clause.
• The substitutions used to produce the empty clause are
those under which the opposite of the negated goal is
true(12.2.5).
Resolution Theorem Proving 3
12.2.1 Resolution Principle(3)
• Prove that “Fido will die.” from the statements
“Fido is a dog.”,
“All dogs are animals.” and
“All animals will die.”
• Changing premises to predicates
• (x) (dog(X) → animal(X))
• dog(fido)
• Modus Ponens and {fido/X}
• animal(fido)
• (Y) (animal(Y) → die(Y))
• Modus Ponens and {fido/Y}
• die(fido)
4
Resolution Theorem Proving
12.2.1 Resolution Principle(4)
q Equivalent Reasoning by Resolution
Ü Convert predicates to clause form
Predicate form Clause form
1. (x) (dog(X) → animal(X)) ¬dog(X) animal(X)
2. dog(fido) dog(fido)
3. (Y) (animal(Y) → die(Y))
¬animal(Y) die(Y)
Ü Negate the conclusion
4. ¬die(fido) ¬die(fido)
Resolution Theorem Proving 5
12.2.1 Resolution Principle(4)
• Equivalent Reasoning by Resolution(continued)
¬dog(X) animal(X) ¬animal(Y) die(Y)
{Y/X}
dog(fido) ¬dog(Y) die(Y)
{fido/Y}
die(fido) ¬die(fido)
Resolution proof for the “dead dog” problem
Resolution Theorem Proving 6
12.2.2 Converting to Clause
Form(1)
q Step 1: Eliminate the logical connectives → and
Ü a b = (a → b) (b → a)
Ü a → b = ¬a b
q Step 2: Reduce the scope of negation
Ü ¬(¬a) = a
Ü ¬(a b) = ¬a ¬b
Ü ¬(a b) = ¬a ¬b
Ü ¬( X) a(X) = ( X) ¬a(X)
Ü ¬( X) b(X) = ( X) ¬b(X)
Resolution Theorem Proving 7
12.2.2 Converting to Clause
Form(2)
q Step 3: Standardize by renaming all variables so
that variables bound by different quantifiers have
unique names.
Ü ( X) a(X) ( X) b(X) = ( X) a(X) ( Y) b(Y)
q Step 4: Move all quantifiers to the left to obtain a
prenex normal form.
q Step 5: Eliminate existential quantifiers by using
skolemization.
Resolution Theorem Proving 8
12.2.2 Converting to Clause
Form(3)
q Step 6: Drop all universal quantifiers
q Step 7: Convert the expression to the conjunction
of disjuncts form
Ü (a b) (c d)
= (a (c d)) (b (c d))
= (a c) (a d) (b c) (b d)
q step 8: Call each conjunct a separate clause
q step 9: Standardize the variables apart again.
Variables are renamed so that no variable
symbol appears in more than one clause.
( X)(a(X) b(X))=( X)a(X) ( Y)b(Y)
Resolution Theorem Proving 9
12.2.2 Converting to Clause
Form(4)
q Skolemization
Ü Skolem constant
Ø ( X)(dog(X)) may be replaced by dog(fido) where the name fido
is picked from the domain of definition of X to represent that
individual X.
Ü Skolem function
Ø If the predicate has more than one argument and the
existentially quantified variable is within the scope of universally
quantified variables, the existential variable must be a function
of those other variables.
Ø ( X)( Y)(mother(X,Y)) ( X)mother(X,m(X))
Ø ( X)( Y)( Z)( W)(foo (X,Y,Z,W))
( X)( Y)( W)(foo(X,Y,f(X,Y),w))
Resolution Theorem Proving 10
12.2.2 Converting to Clause
Form(5)
q Example of Converting Clause Form
( X)([a(X) b(X)] [c(X,I) ( Y)(( Z)[C(Y,Z)] d(X,Y))])
( X)(e(X))
Ü step 1: ( X)(¬[a(X) b(X)] [c(X,I) ( Y)(¬( Z)[c(Y,Z)]
d(X,Y))]) ( x)(e(X))
Ü step 2: ( X)([¬a(X) ¬b(X)] [c(X,I) ( Y)(( Z)[¬c(Y,Z)]
d(X,Y))]) ( x)(e(X))
Ü step 3: ( X)([¬a(X) ¬b(X)] [c(X,I) ( Y)(( Z)[¬c(Y,Z)]
d(X,Y))]) ( W)(e(W))
Ü step 4: ( X)( Y)( Z)( W)( [¬a(X) ¬b(X)] [c(X,I) (¬c(Y,Z)
d(X,Y))]) (e(W))
Ü step 5: ( X)( Z)( W)( [¬a(X) ¬b(X)] [c(X,I) (¬c(f(X),Z)
d(X,f(X)))]) (e(W))
Ü step 6: [¬a(X) ¬b(X)] [c(X,I) (¬c(f(X),Z) d(X,f(X)))]) e(W)
Resolution Theorem Proving 11
12.2.2 Converting to Clause
Form(6)
q Example of Converting Clause Form(continued)
Ü step 7: [가 나] [다 (라 마)] 바
= [가 나 다 바] [가 나 라 마 바]
[¬a(X) ¬b(X) c(X,I) e(W)]
[¬a(X) ¬b(X) ¬c(f(X),Z) d(X,f(X)) e(W)]
Ü step 8: (i) ¬a(X) ¬b(X) c(X,I) e(W)
(ii) ¬a(X) ¬b(X) ¬c(f(X),Z) d(X,f(X)) e(W)
Ü step 9: (i) ¬a(X) ¬b(X) c(X,I) e(W)
(ii) ¬a(U) ¬b(U) ¬c(f(U),Z) d(U,f(U)) e(V)
Resolution Theorem Proving 12
12.2.3 Binary Resolution Proof
Procedure(1)
q Binary Resolution Step
Ü For any two clauses C1 and C2, if there is a literal L1 in C1
that is complementary to a literal L2 in C2, then delete L1 and
L2 from C1 and C2 respectively, and construct the disjunction
of the remaining clauses. The constructed clause is a
resolvent of C1 and C2.
q Examples of Resolution Step
Ü C1=a ¬b, C2=b c
Ø Complementary literals : ¬b,b
Ø Resolvent: a c
Ü C1=¬a b c, C2=¬b d
Ø Complementary literals : b, ¬b
Ø Resolvent : ¬a c d
Resolution Theorem Proving 13
12.2.3 Binary Resolution Proof
Procedure(2)
q Justification of Resolution Step
Ü Theorem
Ø Given two clause C1 and C2, a resolvent C of C1 and C2 is a
logical consequence of C1 and C2.
Ü Proof
Ø Let C1=L C1’, C2=¬L C2’, and C=C1’ C2’, where C1’ and
C2’ are disjunction of literals.
Ø Suppose C1 and C2 are true in an interpretation I.
Ø We want to prove that the resolvent C of C1 and C2 is also true
in I.
Ø Case 1: L is true in I
• Then since C2 = ¬L C2’ is true in I, C2’ must be true in I, and
thus C=C1’ C2’ is true in I.
Ø Case 2: L is false in I
• Then since C1 = L C1’ is true in I, C1’ must be true in I. Thus,
C=C1’ C2’ must be true in I.
Resolution Theorem Proving 14
12.2.3 Binary Resolution Proof
Procedure(3)
q Resolution in Propositional Logic
1. a ← b c a ¬b ¬c
2. b b
3. c ← d e c ¬d ¬e
4. e f e f
5. d ¬f d
¬f
Resolution Theorem Proving 15
12.2.3 Binary Resolution Proof
Procedure(4)
q Resolution in Propositional Logic (continued)
Ü First, the goal to be
¬a a ¬b ¬c
proved, a , is negated
and added to the clause
¬b ¬c b
set.
Ü The derivation of
¬c c ¬d ¬e
indicates that the
database of clauses
e f ¬d ¬e
is inconsistent.
d f ¬d
f ¬f
Resolution Theorem Proving 16
12.2.3 Binary Resolution Proof
Procedure(5)
q Resolution on the predicate calculus
Ü A literal and its negation in parent clauses produce a
resolvent only if they unify under some substitution σ. σ Is
then applied to the resolvent before adding it to the clause
set.
Ü C1 = ¬dog(X) animal(X)
C2 = ¬animal(Y) die(Y)
Resolvent : ¬dog(Y) die(Y) {Y/X}
Resolution Theorem Proving 17
12.2.3 Binary Resolution Proof
Procedure(6)
q “Lucky student”
1. Anyone passing his history exams and winning the lottery is
happy
Ø X(pass(X,history) win(X,lottery) → happy(X))
2. Anyone who studies or is lucky can pass all his exams.
Ø X Y(study(X) lucky(X) → pass(X,Y))
3. John did not study but he is lucky
Ø ¬study(john) lucky(john)
4. Anyone who is lucky wins the lottery.
Ø X(lucky(X) → win(X,lottery))
Resolution Theorem Proving 18
12.2.3 Binary Resolution Proof
Procedure(7)
q Clause forms of “Lucky student”
1. ¬pass(X,history) ¬win(X,lottery) happy(X)
2. ¬study(X) pass(Y,Z)
¬lucky(W) pass(W,V)
3. ¬study(john)
lucky(john)
4. ¬lucky(V) win(V,lottery)
5. Negate the conclusion “John is happy”
¬happy(john)
Resolution Theorem Proving 19
12.2.3 Binary Resolution Proof
Procedure(8)
q Resolution refutation for the “Lucky Student”
problem
¬pass(X, history) ¬win(X,lottery) happy(X) win(U,lottery) ¬lucky(U)
{U/X}
¬pass(U, history) happy(U) ¬lucky(U) ¬happy(john)
{john/U}
lucky(john) ¬pass(john,history) ¬lucky(join)
{}
¬pass(john,history) ¬lucky(V) pass(V,W)
{john/V,history/W}
¬lucky(john) lucky(john)
{}
Resolution Theorem Proving 20
12.2.3 Binary Resolution Proof
Procedure(9)
q “Exciting Life”
1. All people who are not poor and are smart are happy.
Ø X(¬poor(X) smart(X) → happy(X))
2. Those people who read are not stupid.
Ø Y(read(Y) → smart(Y))
Ø {assume X(smart(X) ¬stupid(X))}
3. John can read and is wealthy.
Ø read(john) ¬poor(john)
Ø {assume Y(wealthy(Y) ¬poor(Y))}
4. Happy people have exciting lives.
Ø Z(Happy(Z) → exciting(Z))
5. Negate the conclusion.
Ø “Can anyone be found with an exciting life?”
Ø X(exciting(W))
Resolution Theorem Proving 21
12.2.3 Binary Resolution Proof
Procedure(10)
q Clause forms of “exciting life”
1. poor(X) ¬smart(X) happy(X)
2. ¬read(Y) smart(Y)
3. read(john)
¬poor(john)
4. ¬happy(Z) exciting(Z)
5. ¬exciting(W)
Resolution Theorem Proving 22
12.2.3 Binary Resolution Proof
Procedure(11)
q Resolution refutation for the “exciting life”
¬exciting(W) ¬happy(Z) exciting(Z)
{Z/W}
¬happy(Z) poor(X) ¬smart(X) happy(X)
{X/Z}
poor(X) ¬smart(X) ¬read(Y) smart(Y)
{Y/X}
¬poor(john) poor(Y) ¬read(Y)
{john/Y}
¬read(john) read(john)
{}
Resolution Theorem Proving 23
12.2.3 Binary Resolution Proof
Procedure(12)
q Another resolution refutation for “exciting life”
¬happy(Z) exciting(Z) poor(X) ¬smart(X) happy(X)
{Z/X}
¬read(Y) smart(Y) exciting(Z) poor(Z) ¬smart(Z)
{Y/Z}
¬read(Y) exciting(Y) poor(Y) ¬poor(john)
{john/Y}
read(john) ¬read(john) exciting(john)
{}
exciting(john) ¬exciting(W)
{}
Resolution Theorem Proving 24
12.2.4 Strategies for Resolution(1)
q Order of clause combination is important
Ü N clauses → N2 ways of combinations or checking to see
whether they can be combined
Ü Search heuristics are very important in resolution proof
procedures
q Strategies
Ü Breadth-First Strategy
Ü Set of Support Strategy
Ü Unit Preference Strategy
Ü Linear Input Form Strategy
Resolution Theorem Proving 25
12.2.4 Strategies for Resolution(2)
q Breadth-First Strategy
Ü First round: each clause is compared for resolution with
every other clause in the clause space.
Ü Second round: generate the clauses by resolving the
clauses produced at the first round with all the original
clauses.
Ü Nth round: generate the clauses by resolving all clauses at
level n-1 against the elements of the original clause set and
all clauses previously produced.
Ü Characteristics
Ø it guarantees finding the shortest solution path because it
generates every search state for each level before going any
deeper.
Ø It is a complete strategy in that it is guaranteed to find a
refutation if one exists.
Ü Figure 12.8 (p. 530)
Resolution Theorem Proving 26
12.2.4 Strategies for Resolution(3)
q Set of Support Strategy
Ü Specify a subset(T) of a set of input clauses(S) called the set
of support such that S-T is satisfiable.
Ü Require that resolvents in each resolution have an ancestor
in the set of support(T).
Ü Based on the insight that the negation of what we want to
prove true is going to be responsible for causing the clause
space to be contradictory.
Ü Forces resolutions between clauses of which at least one is
either the negated goal clause or a clause produced by
resolutions on the negated goal.
Ü Figure 12.6 (p. 528)
Resolution Theorem Proving 27
12.2.4 Strategies for Resolution(4)
q Unit Preference Strategy
Ü Try to produce a resultant clause that has fewer literals than
the parent clauses.
Ü Resolving with a clause of one literal, called a unit clause,
will guarantee that the resolvent is smaller than the largest
parent clause.
Ü Figure 12.9 (p. 531)
Resolution Theorem Proving 28
12.2.4 Strategies for Resolution(5)
q Linear Input Form Strategy
Ü a direct use of the negated goal and the original axioms
Ü Take the negated goal and resolve it with one of the axioms.
The result is then resolved with one of the axioms to get
another new clause. The new clause is again resolved with
one of the axioms.
Ü Try to resolve the most recently obtained clause with the
original axioms.
Resolution Theorem Proving 29
12.2.5 Answer Extraction from
Resolution Refutations(1)
q Extract the correct answer by retaining
information on the unification substitutions made
in the resolution refutation.
Ü Retain the original conclusion that was to be proved.
Ü Introduce each unification made in the resolution process
into the conclusion.
q Example of answer extraction
Ü Fido goes wherever John goes.
Ø at(john, X) → at(fido, X)
Ü John is at the library.
Ø at(john, library)
Ü Negate the conclusion “Where is Fido?”
Ø ¬at(fido, Z)
Resolution Theorem Proving 30
12.2.5 Answer Extraction from
Resolution Refutations(2)
q Example of answer extraction(continued)
at(fido, Z)
¬at(fido, Z) ¬at(john, X) at(fido, X)
{X/Z}
at(fido, X)
¬at(john, X) at(john, library)
{library/X}
at(fido, library)
Resolution Theorem Proving 31
12.2.5 Answer Extraction from
Resolution Refutations(3)
q “Exciting Life”
exciting(W)
¬exciting(W) ¬happy(Z) exciting(Z)
{Z/W} {Z/W}
exciting(Z)
¬happy(Z) poor(X) ¬smart(X) happy(X)
{X/Z} {X/Z}
poor(X) ¬smart(X)
¬read(Y) smart(Y)
exciting(X)
{Y/X}
{Y/X}
¬poor(john) poor(Y) ¬read(Y)
exciting(Y)
{john/Y} {john/Y}
exciting(john)
¬read(john) read(john)
{} {}
exciting(john)
Resolution Theorem Proving 32
