Pregnancy period (before 1956)

Formative period (1956-1970)

Dark period (1966-1974)

Knowledge application period (1970-1988)

Integrated development period (1986 to present)

Symbolism Mathematical Logic Newell Simon Nelson Functional Simulation Method

Connectionism Biomimicry Carlock Pitts Hopfield Rummelhardt Structural simulation methods

Behaviorism Cybernetics Brooks Hexapod walking robot Behavior simulation method

Short-term goal: Build intelligent computers to replace some of human intellectual activities

Long-term goal: Use automata to imitate human thinking activities and intellectual functions

Problem solving and gaming

Logical reasoning and theorem proving

computational intelligence

Distributed Artificial Intelligence and Reality

automatic programming

expert system

machine learning

natural language understanding

Robotics

pattern recognition

Neural Networks

machine vision

intelligent control

authenticity

relativity

incompleteness

Uncertainty

representability

Storeability Deliverability Processability

compatibility

state space representation

problem reduction representation

predicate logic representation

semantic network representation

frame representation

ontology technology

process representation

General search strategies for graphs

blind search

heuristic search

American mathematician Robinson

Starting point: To prove that a proposition is true, we can prove that its negative proposition is false.

Atomic formula

Text: An atomic common formula and its negation

Clause: a well-formed formula consisting of a disjunction of words

Digestion: Decompose and simplify the predicate calculus formula, and eliminate some symbols to derive clauses.

The resolution is the logical conclusion of the parent clause

Resolution can only be performed between formulas containing negation and disjunctive connectives.

Obtaining the set of clauses

Elimination inference rules: C1＝LVC1＇C2＝~LⅤC2＇→C12＝C1VC2

Replacement and Unification

Total database

production rules

Control Strategy

nonmonotonic reasoning

random uncertainty

fuzzy uncertainty

incompleteness

Inconsistency

uncertainty of knowledge

uncertainty about the evidence

uncertainty of conclusion

Bayesian formula

LS is called the sufficiency measure, which is used to indicate the degree of support of E for H LS=P(E/H)/P(E/~H) Its value is given by domain experts

LN is called the necessity measure, which is used to indicate the degree of support of E for H LN=P(~E/H)/P(~E/~H)=[1-P(E/H)]/[1-P(E/~H)] Its value is given by domain experts

Probability function O(x)=P(x)/1-P(x)

The corresponding relationship between the initial credibility C(E/S) and the probability P(E/S) is as follows