AI

First Order Logic

• First Order Logic (FOL) handles limitations of propositional logic by allowing relations about objects and functions on objects.
• FOL allows statements to be true, false, or unknown, extending propositional logic which only handled facts.
• FOL Constructs:
  • Constants: Represent objects, for example, "john", "apples"
  • Predicates: Represent properties and relations, for example, "likes(john, apples)"
  • Functions: Transform objects, for example, "likes(john, fruit_of(apple_tree))"
  • Variables: Represent any object, for example, "likes(X, apples)"
  • Quantifiers: Qualify variable values
    • Universal Quantifier (): True for all objects, for example, "X.likes(X, apples)"
    • Existential Quantifier (): Exists at least one object, for example, "X.likes(X, apples)"

Prolog

• Prolog (Programming in Logic) is a declarative programming tool used to write AI programs.
• Prolog is based on FOL syntax and consists of:
  • Facts: Truths about the world
  • Rules: Logical deductions

Production Rules

• Production rules are sets of "if condition then action" pairs.
• The agent uses a "match-resolve-act" cycle:
  • Match: Checks if each rule's condition holds.
  • Resolve: Chooses a rule from a potential set of applicable rules.
  • Act: Executes the action associated with the chosen rule.
• Working memory:
  • Rules can write knowledge into working memory, which can trigger other rules.

Semantic Networks

• Semantic networks represent objects as nodes.
• These networks are hierarchical and show relationships between objects.
• Inheritance of properties is similar to object-oriented programming (OOP).

Knowledge Representation

• When dealing with knowledge, there are two key aspects:
• Representing the knowledge space
• Implementing reasoning processes.
• The goal is to represent the world logically to allow intelligent agents to make decisions.
• Knowledge representation is the process of storing and processing knowledge efficiently.
• Knowledge is crucial for intelligent systems to make decisions.
• An intelligent agent utilizes a knowledge base (KB) to represent facts about the world.

What is Knowledge

• Data: Primitive verifiable facts, often voluminous and frequently changing, reflecting the current state of the world.
• Information: Interpreted data.
• Knowledge: Relationships between sets of data used for further deduction. Knowledge is general and provides information about the behavior of abstract models of the world.

General Knowledge Representations Schemes

• Logic-based representation: First-order predicate logic (FOL)
• Procedural representation: Production rules
• Network representation: Semantic networks, conceptual graphs
• Structural representation: Scripts, frames, objects

Propositional Logic

• Proposition: A statement that is either true or false.
• Logical Operators: Combine propositions
  • Conjunction (and) ()
  • Disjunction (or) ()
  • Negation (not) ()
  • Implication ( )
  • Equivalence ()

Propositional Logic Limitations

• Can only handle true or false values.
• Entire propositions are represented as single symbols.
• It does not allow representation of objects with properties (height, width, etc.), and relations between them.
• It does not easily support changes to the knowledge base.
• Lack of expressiveness and inability to talk about specifics.