Recap of Lecture 5 and Homework Check
LAST LECTURE · 5 MIN
Knowledge
Three methods
Weakness
Data → information → knowledge. A KB agent uses TELL and ASK.
Semantic networks, frames, production rules (forward chaining).
None of them can say NOT, OR, ALL, SOME exactly. Logic can.
Today: the same tubewell knowledge written in logic, and a computer that PROVES the answer (S22).
Today's Plan (90 minutes)
ROADMAP
0-5
Recap of Lecture 5 and homework check
S02
5-8
Plan and learning outcomes
S03-S04
8-20
Module 1: Logic and propositions
S05-S08
20-40
Module 2: Five connectives and translating English
S09-S13
40-58
Module 3: Truth tables; tautology, contradiction, contingency
S14-S18
58-74
Module 4: Equivalences, models and entailment
S19-S23
74-84
Class activity: load-shedding logic
S24
84-90
Mistakes, recap quiz, homework
S25-S27
Learning Outcomes (SLOs)
BY THE END OF THIS LECTURE YOU WILL BE ABLE TO
SLO 6.1
Identify propositions and differentiate atomic from compound propositions
→ CLO2 · C2 Understand
SLO 6.2
Translate English sentences into propositional logic using the five connectives
→ CLO2 · C3 Apply
SLO 6.3
Build a truth table for a compound proposition using operator precedence
→ CLO2 · C3 Apply
SLO 6.4
Classify formulas as tautology, contradiction or contingency and prove equivalences
→ CLO2 · C3 Apply
SLO 6.5
Decide whether a knowledge base entails a sentence by model checking
→ CLO2 · C3 Apply
- Course Learning Outcome(s) served
- CLO2: Demonstrate first-order predicate calculus, logical reasoning and problem-solving using Prolog language.
Slide 5
MODULE 1
Logic and Propositions
Sentences that are either true or false - nothing in between
What Is Logic and Why Does AI Need It?
MODULE 1 · LOGIC
Syntax
Semantics
Also called
- Which sentences are allowed (the grammar).
- 'P ∧ Q' is allowed; 'P ∧ ∧ Q' is not.
What a sentence means: in which situations it is TRUE or FALSE.
Propositional calculus, Boolean logic, sentential logic. It works with 1 (true) and 0 (false).
Logic: an exact language for knowledge, with clear grammar (syntax) and clear meaning (semantics).
Propositions: True or False, Not Both
MODULE 1 · PROPOSITIONS
Proposition: a declarative sentence that is either TRUE or FALSE, but not both.
Atomic and Compound Propositions
MODULE 1 · PROPOSITIONS
Atomic
Compound
- One simple statement, one symbol.
- L: There is load-shedding.
- U: The UPS is charged.
- F: The fan is running.
- Atoms joined by connectives.
- 'There is load-shedding and the UPS is charged' → L ∧ U
- 'If L and U then the fan runs' → (L ∧ U) → F
Always: write a KEY first (which letter means which statement), then the formula.
Key fact: the truth of a compound proposition depends only on the truth of its atoms.
Slide 9
MODULE 2
Logical Connectives
Five operators that join simple sentences into compound ones
The Five Connectives
MODULE 2 · CONNECTIVES
Careful: OR in logic is INCLUSIVE - P ∨ Q is also true when both P and Q are true.
Understanding Implication: If P Then Q
MODULE 2 · CONNECTIVES
Rule: P → Q is FALSE only when P is true and Q is false. In every other row it is true.
Biconditional and Operator Precedence
MODULE 2 · CONNECTIVES
- P ↔ Q is true when both sides have the SAME value.
- P ↔ Q ≡ (P → Q) ∧ (Q → P)
- Example: a number is even ↔ it is divisible by 2.
Precedence (highest first) - like BODMAS:
1. ¬
2. ∧
3. ∨
4. →
5. ↔
Example: ¬P ∧ Q ∨ R → S means (((¬P) ∧ Q) ∨ R) → S. Exam tip: always write the brackets.
Translating English into Propositional Logic
MODULE 2 · TRANSLATION
Key: L = there is load-shedding U = the UPS is charged F = the fan runs S = Ali studies E = there is an exam tomorrow
Watch: 'P only if Q' is P → Q (not Q → P). 'Q if P' is also P → Q.
Slide 14
MODULE 3
Truth Tables
Check every possible situation - a computer's way of being sure
Truth Table of the Five Connectives
MODULE 3 · TRUTH TABLES
REAL OUTPUT: L06/TRUTH_TABLE.PY
AND
OR
- Truth table of the five connectives
- P Q | ¬P | P∧Q | P∨Q | P→Q | P↔Q
- T T | F | T | T | T | T
- T F | F | F | T | F | F
- F T | T | F | T | T | F
- F F | T | F | F | T | T
P ∧ Q is true in ONE row only (both true).
P ∨ Q is false in ONE row only (both false).
IF-THEN
IFF
P → Q is false in ONE row only (T → F).
P ↔ Q is true when both are the SAME.
Rows: n atoms → 2ⁿ rows. Always list them in the order TT, TF, FT, FF.
Building a Truth Table Step by Step
MODULE 3 · TRUTH TABLES
- Step 1: list the atoms (P, Q) and all 2ⁿ rows.
- Step 2: add one column for each sub-formula, inside brackets first.
- Step 3: the last column is the whole formula.
- Story: P = fault is in the fuse, Q = fault is in the wiring. 'Fuse or wiring; not the fuse; so the wiring.'
REAL OUTPUT: ((P ∨ Q) ∧ ¬P) → Q
- Step by step: ((P ∨ Q) ∧ ¬P) → Q
- P Q | P∨Q | ¬P | (P∨Q)∧¬P | whole
- T T | T | F | F | T
- T F | T | F | F | T
- F T | T | T | T | T
- F F | F | T | F | T
Result: the last column is ALL TRUE - this way of reasoning is always valid (disjunctive syllogism).
Demo: A Truth-Table Maker in Python
MODULE 3 · TRUTH TABLES
L06/PL.PY (CORE)
A model
- from itertools import product
- def implies(a, b): return (not a) or b
- def iff(a, b): return a == b
- def models(names): # all 2^n rows
- for values in product([True, False], repeat=len(names)):
- yield dict(zip(names, values))
- def classify(names, f):
- vals = [f(m) for m in models(names)]
- if all(vals): return "tautology (valid)"
- if not any(vals): return "contradiction (unsatisfiable)"
- return "contingency (satisfiable)"
- One row of the table = one possible world:
- {'P': True, 'Q': False}
A formula
- A function of the model:
- lambda m: implies(m['P'], m['Q'])
Tautology, Contradiction and Contingency
MODULE 3 · TRUTH TABLES
REAL OUTPUT: L06/CLASSIFY.PY
Why AI cares: a valid formula is a safe reasoning pattern; a contradiction in a KB means wrong facts.
- P ∨ ¬P tautology (valid)
- P ∧ ¬P contradiction (unsatisfiable)
- P → Q contingency (satisfiable)
- ((P ∨ Q) ∧ ¬P) → Q tautology (valid)
- (P → Q) ∧ P ∧ ¬Q contradiction (unsatisfiable)
Slide 19
MODULE 4
Equivalence and Entailment
When do two formulas mean the same? When does knowledge prove a conclusion?
Important Logical Equivalences
MODULE 4 · EQUIVALENCE
Equivalent (≡): identical truth-table columns in EVERY row.
Proving Equivalence with a Computer
MODULE 4 · EQUIVALENCE
REAL OUTPUT: L06/CLASSIFY.PY (EQUIVALENCE CHECKS)
Converse trap
- P → Q ≡ ¬P ∨ Q True
- P → Q ≡ ¬Q → ¬P (contrapositive) True
- P → Q ≡ Q → P (converse) False
- ¬(P ∧ Q) ≡ ¬P ∨ ¬Q (De Morgan) True
- ¬(P ∨ Q) ≡ ¬P ∧ ¬Q (De Morgan) True
- ¬(P ∧ Q) ≡ ¬P ∧ ¬Q (wrong!) False
'If it rains, the road is wet.' Does a wet road mean it rained? NO - a water tanker passed by. P → Q is not Q → P.
By hand: two rows are enough to show P → Q and Q → P are NOT equivalent.
Models and Entailment (KB ⊨ α)
MODULE 4 · ENTAILMENT
EXAMPLE 1: KB = {R → W, R}
EXAMPLE 2: TUBEWELL KB (L05 HOMEWORK)
- R W | R→W R | KB | W
- T T | T T | T | T
- T F | F T | F | F
- F T | T F | F | T
- F F | T F | F | F
- KB entails W: True
- KB entails G (start diesel engine) : True
- KB entails I (irrigation needed) : True
- KB entails T (start tubewell) : False
- KB entails ¬T (tubewell off) : False
- Models checked: 64 | models where KB is true: 2
Tubewell: KB ⊨ G (start diesel engine) - proved in all 64 worlds. Same answer as forward chaining in L05.
Strengths and Limits of Propositional Logic
MODULE 4 · LIMITS
Lecture 7: predicate logic
Lecture 8: resolution
- ∀x Student(x) → NeedsAttendance(x, 75)
- Predicates, variables, ALL (∀) and SOME (∃).
Proves entailment without writing every row of the truth table.
Activity: Load-Shedding Logic
CLASS ACTIVITY · PAIRS · 10 MIN
Pairs. Key: L = load-shedding, U = UPS is charged, F = fan runs, S = Ali studies. Write and check your answers.
Translate: 'If there is load-shedding and the UPS is charged, the fan runs.'
Translate: 'There is no load-shedding, but the UPS is not charged.'
Translate: 'The fan runs only if the UPS is charged or there is no load-shedding.'
Evaluate sentence 1 when L = T, U = F, F = F.
Translate: 'Ali studies unless there is load-shedding.'
Is sentence 1 equivalent to its converse F → (L ∧ U)? Give one row that decides.
Common Mistakes
WATCH OUT
WRONG IDEA
CORRECT IDEA
✗ Questions and commands are propositions.
✓ Only declarative sentences with a true/false value are propositions.
✗ P → Q is false when P is false.
✓ P → Q is false ONLY when P is true and Q is false.
✗ P → Q means the same as Q → P.
✓ The converse is different; only the contrapositive ¬Q → ¬P is equivalent.
✗ ¬(P ∧ Q) ≡ ¬P ∧ ¬Q
✓ De Morgan: ¬(P ∧ Q) ≡ ¬P ∨ ¬Q (the ∧ flips to ∨).
✗ 'P only if Q' means Q → P.
✓ 'P only if Q' means P → Q.
Recap Quiz
CHECK YOUR UNDERSTANDING · 5 MIN
- Q1. Which is a proposition?
- (a) Close the door (b) Is it raining? (c) Lahore is in Punjab (d) x + 1 = 3
- Q2. P → Q is false when:
- (a) P = T, Q = T (b) P = T, Q = F (c) P = F, Q = T (d) P = F, Q = F
- Q3. Highest precedence connective:
- (a) ∧ (b) ∨ (c) ¬ (d) →
- Q4. P ∨ ¬P is a:
- (a) Contradiction (b) Contingency (c) Tautology (d) Not a formula
- Q5. ¬(P ∨ Q) is equivalent to:
- (a) ¬P ∨ ¬Q (b) ¬P ∧ ¬Q (c) P ∧ Q (d) ¬P → Q
- Q6. A truth table with 3 atoms has how many rows?
- (a) 3 (b) 6 (c) 8 (d) 9
Homework and Next Lecture
BEFORE THE NEXT CLASS
- Task 1: Key: R = it rains, C = canal water is available, W = farmer waters the wheat, Y = yield is good. Translate: (a) If it rains, the farmer does not water the wheat. (b) The yield is good only if the wheat is watered or it rains. (c) The farmer waters the wheat if and only if canal water is available and it does not rain. (d) Neither rain nor canal water is available. (e) The farmer waters unless it rains.
- Task 2: Make truth tables for (P → Q) ↔ (¬Q → ¬P) and (P → Q) ↔ (Q → P). Classify each.
- Task 3: A = Ali's attendance is 75%+, E = Ali is eligible. Show with a truth table that {A → E, ¬E} ⊨ ¬A.
- Read: Russell & Norvig 7.4; Luger 2.1.
NEXT LECTURE
- Lecture 7:
- Predicate (First-Order) Logic