Please enable JavaScript.
Coggle requires JavaScript to display documents.
LEARNING UNIT 5: ARTIFICIAL INTELLIGENCE - Coggle Diagram
LEARNING UNIT 5: ARTIFICIAL INTELLIGENCE
Historical Background
Turing Machine
able to solve problem
solving using application of a set of precise rules & instuctions
Turing's Proposal (Automated Computing Engine 1946)
Essential aspects of computer
Convert binary form of calculations to decimal form of input & output
Logical Control
Central arithmetic part to carry out fundamental arithmetic
Erasable memory
Advantages
Avoid human error
carry out complicated task
Speed
The Turing Test
Electronic Numeral Integrator & Calculator ( ENIAC )
Von Neumann's mathematical ability contributed to the design of ENIAC ( Primary compuiting engine for the US army
During war, interest in speeding up computation was underlined by the need to be able to calculate the speed and trajectory of bombs and missiles
To calculate artillery tables making it possible to aim artillery at specific targets
Electronic Discrete Variable Automatic Computer ( EDVAC)
Used at the Ballistic Reserach Laboratories at Aberdeen, Maryland
Analytical Engine ( Giant Calculating Machine)
The idea of an advanced calculating machine to calculate & print mathematical tables from Babbage
Never built except for a trial piece before Babbage' death
Inspired by the idea by Howard Aiken - led to the development of Mark1
The Jacquard Loom:
Ancestor of Computer
A system that allows any matter to be automatically woven onto fabrics
COPYING HUMAN LIMITATIONS
Moving beyond human
physical limitations
Expanders & Enhancers of
Human ‘Brainpower’
INTELLIGENT MACHINES: EARLY PROGRAMS
HUMAN LOGIC
Human exercise reasoning when solving problems
Logical steps
Vast & complex world knowledge as background
Program computer so that it can engage in reasoning like that of human, the vast, complex world must be simplified
LOGIC THEORIST PROGRAM
"the first artificial intelligence program"
by (
Newell, Shaw & Simon, 1956
) the first program deliberately engineered to mimic the problem-solving skills of human beings.
THE GENERAL PROBLEM SOLVER (
GPS)
certain types of problem, whose solutions lend themselves to a very explicit and clear steps can be formulated for the GPS
searching among and following a series of elementary reasoning steps, similar to those carried out by human.
researchers soon realized that developing such general purpose systems was too difficult focus on systems for
limited domains
Expert systems
depend on human expert assistance in our lives
doctors
attorneys
automobile mechanics
computer repairmen
can expert assistance be given by computer programs?
prospector : identify sites for drilling of oil
Dendral : suggest the chemical structure of unknown compounds
MYCIN : choose appropriate antibiotics for patients with severe bacterial infections
what is an expert system?
systems which encode human expertise in limited domains
a computer program designed to hold the accumulated knowledge of one or more domain experts.
The Applications
PROSPECTOR
used by geologists to identify sites for drilling or mining
PUFF
medical system for diagnosis of respiratory conditions
DESIGN ADVISOR
gives advice to designers of chips processor
MYCIN
medical system for diagnosing blood disorders
LITHIAN
gives advice to archaeologists examining stone tools
DENDRAL
used to identify the structure of chemical compounds. first used in 1965
THE ISSUES
WHY THE NEED?
not always available
can be used anywhere, any time
human experts are not 100% reliable or consistent
may not good at explaining decisions
cost effective
PROBLEM?
limited edition
systems are not always up to date, and don't learn
no "
common sense
"
needed to setup and maintain system
BEGINNER
ANCIENT HISTORY
The intellectual roots of AI, and the concept of intelligent machines, may be found in Greek mythology.
Invention of human-like artifacts
FRANKESTEIN
Original story, published by Mary Shelley, in 1818, describes the attempt of a true scientist, Victor Frankenstein, to create life.
Modern History
1956 - John McCarthy coined the term "artificial intelligence" as the topic of the Dartmouth Conference, the first conference devoted to the subject.
What is Artificial Intelligence?
The study of how to make computers/ machines do things, which, at the moment, people are better.
Machines that perform tasks that seem to require intelligence
(or some facet of intelligence) when performed by humans.
Smell
Intuition
Speech Recognition
Inferencing
Learning Skill
Decision Making
Focus of Artificial Intelligence
The identification of the significance, interpretation, or explanation of certain data or information (i.e. the ability to employ knowledge)
The ability to generate new ideas or conceive new perspectives on existing ideas – producing ideas which are original and potentially useful
The ability of drawing conclusions appropriate to the situation in hand
The process of acquiring knowledge, skills, experience or values by study, experience or training
Turing Test - by Alan Turing (1950)
Designed to provide a satisfactory operational definition of intelligence
Intelligent behavior = the ability to achieve human-level performance in all cognitive tasks, sufficient to fool an interrogator
Turing Test is meant to determine if a computer program has intelligence
any machine whose output was indistinguishable from a human's output
Turing Test
Judge/interrogator conversations with human & machine
Judge cannot distinguish the machine from the human on the basis of the conversation
Cognitive skills- required of Intelligence Machine
knowledge representation
store information before or during interrogation
Automated Reasoning
stores information to answer questions & draw conclution
Natural language processing
enable to communicate in English
Machine Learning
adapt new circumstances, detect & extrapolate
Understanding AI
system thinking & acting humanly
thought process & reasoning
sufficiently precise theory of human mind, possible to express the theory (computer program)
the real of cognitive science
cognitive scientists- reconstruct theories of hpw the human mind works.
create a model of intelligent human behavior
stimulate the model on a machine.
get inside human minds
Introspection
Psychological experiments
behavior
AI programs have to interact with people, behave according to certain normal conventions of human interaction
Should pass the Turing Test
Do what humans do, regardless of whether it is done the same way as humans
system thinking & acting rationally
thought process & reasoning
System that thinks rationally = Intelligent Machine
They do not think.
They simply calculate.
humans do not always
base their thinking on correct premises
follow steps in logical reasoning
rules : Laws of thought
behavior
rational agent makes
firm
machine
typically a person
software
decisions
An agent is just something
that perceives and acts.
perceives environment through sensors
act upon environemnt through effectors
acting so as to achieve one's goals, given one's beliefs.
agent & its environment
properties:
interacts with other agents plus the environment
reactive to the environment
autonomous
pro-active (goal-directed)
AI strives to build intelligent enities
key issues
artificial intelligent entity
tasks that are larger and complicated
ROBOTICS
Contains sensors, control systems, manipulators, power suppliers & software all working together to perform a task
Machine to perform specifics functions
Robotic: Essential Features
2) Movement
rolling on wheels
walking on legs
propelling by thrusters
3) Energy
solar powered
battery powered
electrically powered
1) Sensing (sensors) = aware of environment
Robot Vision = light sensors
touch & pressure sensors
Chemical sensors
hearing & sonar sensors
taste sensors
4) intelligence
programming = so that it knows what is to do
Robot Vision
Use television camera
Camera that can swivel
comprehend sequences of images
Cog project
Building a Humanoid = resembling human
Human intelligence more complex than that of simple life forms
Insight on human intelligence
Intelligence is achieved over the course
Sensory perception is processed simultaneously
Interaction with world change over time
Perdicted on the notion that these attributes of human intelligence
MACHINE LEARNING (But can machine learn)
Connectionism & Neural Network
Expert System
:
Desribes intelligent behaviour based on the knowledge of human domain expert
No learning mechanism
Artificial Neural Network
:
Immitate intelligent behaviour based on the working of the biological neural network
How to differ neural network from biological neural networks ?
Biological Neural Network
:
Neurons : nerve cells, specialized to received and transmit information in the nervous system
Artificial Neuron Network
:
A system based on the operation of biological neural networks
An emulation of biological neural system
ANN Underlying Concept
:
Model the human brain
Consists of
nodes
, connected by
input
and
output
links
Idea based on :
Links (connected between nodes)
Weights (numeric value)
Network (for activation)
ANN incorporates the above-stated learning properties :
Learning by repetition
Forgetting from disuse
Repetition
: more repetition, more conductance
Unpredictability
: include a probability of firing (a degree of randomness) into the neural net design
Associative learning
: a nodes that fires causes the firing o another, which connects with the conducting link
PERCEPTRON
:
stored information represents experience
a neural net that
learns from experience
machine could be capable of learning
could be trained to produce a desired output for each input pattern
Natural Language Processing
Making computers talk
Systems that translate ordinary human natural language into language computers can understand and act on.
The complexity of programming a computer to "do" human language reflects the complexity of human language
Attempt at NLP: ELIZA (1965) by Joseph Weizenbaum
A program simulated a conversation between a patient and a psychotherapist by using a person's responses to shape the computer's replies
Some people actually mistook her for human
Understanding Language
Parsers
An attempt to address the limitations of programs such as ELIZA
Programs that assign grammatical categories to the words in sentence and group them into phrases
Based on the contributions of contemporary linguistics
Semantic Information Processing
Proposed by Terry Winograd
To model the language understanding process in a computer, we need a programme which combines grammar, semantics, and reasoning in an intimate way
SHRDLU
Created by Terry Winograd
A simple virtual/simulated robot that carries out command given by human
Has natural language processing capabilities based on the expert system ideas
Contains
A dictionary for synatx, semantics, and grammar
A parser
Set of programmes to control responses
Script Applier Mechanism (SAM)
Created by Roger Schank
A computer program designed to understand stories that rely heavily on scripts
Story "Understander" - Make inferences about stories
Deal only with highly restricted scripts
No rival to human intelligence
Attempts made to develop programs that can carry out some cognitive functions of the Human Brain
The programs developed so far are restricted
Human Problem Solving = All background knowledge used, not restricted or predetermined
Computers do not have the amount of knowledge necessary to solve the types of problems human faced everyday
Cannot learn something new & cannot make connections from the experience gained
