M366: Natural and artificial intelligence

1
M366 Natural and artificial intelligence

• 8 credits course, one semester
• Pre-requisite course M263
• Two TMAs (20), one MTA(30) and one final exam
  (50)
• Like any other AOU course, to pass the course you
  have to get
• A Minimum of 40 on the CA (TMA and MTA)
• A Minimum of 40 on the final exam
• A Minimum of 50 for the average of the CA and
  the final

2
Course Structure

• The course is divided into six blocks
• A total of 17 units
• Block 1 intelligent machines
• Unit 1 Machines, minds and computers
• Block 2 Symbolic intelligence
• Unit 1 Fundamentals of symbolic AI
• Unit 2 Search
• Unit 3 Symbolic AI in the world
• Unit 4 Has symbolic AI failed

3
Course Structure the units

• Block 3 Natural intelligence
• Unit 1 Natural intelligence
• Unit 2 Mechanism of natural intelligence
• Unit 3 Interaction and emergence in swarms
• Unit 4 Interaction, emergence, adaptation and
  selection in individuals
• Block 4 Neural networks
• Unit 1 Mechanism
• Unit 2 Layers and learning
• Unit 3 Unsupervised learning in layers and
  lattices
• Unit 4 Its about time recurrence, dynamics and
  chaos

4
Course Structure the units
4

• Block 5 Evolutionary computation
• Unit 1 Unleashing the gene genie, an introduction
  to evolutionary algorithms
• Unit 2 Genetic algorithms
• Unit 3 Artificial evolution
• Block 6 Reflections
• Unit 1 Intelligence, mind and consciousness

5
Block I, Unit 1 Machines, minds and computers

• This unit has two main aims
• Reviewing the development of human thinking about
  machines and our mental ability
• Presenting historical and technical issues that
  lead to Cybernetics and Symbolic AI

6
Block I, Unit 1 Machines, minds and computers
6

• This unit focuses on
• Machines
• Minds
• Artificial intelligence
• Computers

7
Block I, Unit 1 Machines
7

• The early beginning Hephaestus (god of fire) in
  the old Greek, created Talos, a gigantic
  mechanical man of bronze, guardian of Crete.
  (Iliad, XVIII)
• Automata (around 1495), Leonardo da Vinci
  constructed an automaton in the form of armored
  man capable of moving its arms and simulating
  speech.
• Vaucansons duck (1800s)
• Game-playing automata the Turk (1770), Deep blue
  of IBM 1997)
• Robots

8
Block I, Unit 1 Machines
8

• Why build such artificial entities?
• What sort of thing did people think these
  entities actually were?
• What has been the public attitude to the idea of
  artificial creatures?
• Inspired by myths and early creatures, mechanical
  pictures start to appear by thinkers of the 17th
  and the 18th centuries

9
Block I, Unit 1 Machines
9

• An intellect which at a certain moment would know
  all forces that set nature in motion, and all
  positions of all items of which nature is
  composed, if this intellect were also vast enough
  to submit these data to analysis, it would
  embrace in a single formula the movements of the
  greatest bodies of the universe and those of the
  tiniest atom for such an intellect nothing would
  be uncertain and the future just like the past
  would be present before its eyes.
• Source Laplace, Celestial Mechanics
  (17991825)

10
Block I, Unit 1 Minds
10

• What is mind ?
• Mind and body Debate between monist and dualist
• Thomas Hobbes (1588-1679)
• The world consists only of particles of matter in
  motion.
• Bodies and minds are also just particles of
  matter in motion. Their motions are caused, in
  part, by the effects of the movements of
  particles outside the body, which press on the
  senses, causing particles in our minds to move in
  sympathy.
• The particles in our minds form parcels that is,
  symbols representing concepts such as number,
  time, names, and so on.
• Thought amounts to a form of computation, in
  which these mental symbols are added, subtracted,
  etc., in processes similar to those of arithmetic.

11
Block I, Unit 1 Cybernetics
11

• Cybernetics definitions
• "a science concerned with the study of systems of
  any nature which are capable of receiving,
  storing, and processing information so as to use
  it for control"-A.N. Kolmogorov
• "Cybernetique the art of growing"--A.M. Ampere
• "the science of control and communication in the
  animal and the machine"-Norbert Wiener
• "the art of securing efficient operation"-L.
  Couffignal
• "the art of steersmanship" "deals with all forms
  of behavior in so far as they are regular, or
  determinate, or reproducible" "stands to the
  real machine-electronic, mechanical, neural, or
  economic-much as geometry stands to a real object
  in our terrestrial space" "offers a method for
  the scientific treatment of the system in which
  complexity is outstanding and too important to be
  ignored"-W. Ross Ashby

12
Block I, Unit 1 Cybernetics
12

• Cybernetics definitions
• a branch of mathematics dealing with problems of
  control, recursiveness, and information"-Gregory
  Bateson
• "the science of effective organization"-Stafford
  Beer
• "the art and science of manipulating defensible
  metaphors"-Gordon Pask
• "Should one name one central concept, a first
  principle, of cybernetics, it would be
  circularity."-Heinz von Foerster
• "a way of thinking"-Ernst von Glasersfeld
• "the science and art of understanding"-Humberto
  Maturana
• "Cybernetics when I reflect on the dynamics of
  observed systems and on the dynamics of the
  observer-whence 'creative cybernetics' when I
  project the dynamics of a system I would like to
  observe"-from announcement of 1987 ASC conference
  in Urbana-Champaign, Illinois
• "the ability to cure all temporary truth of
  eternal triteness"-Herbert Brun
• source GWU

13
Block I, Unit 1 Cybernetics
13

• cybernetics attempts to find the common elements
  in the functioning of automatic machines and of
  the human nervous system, and to develop a theory
  that will cover the entire field of control and
  communication in machines and in living
  organisms.

14
Block I, Unit 1 Symbolic AI
14

• The goal is to construct machines that have the
  following features
• Use of language
• Forming and using concepts
• Complex problem-solving, such as playing chess
• Learning
• Creativity

15
Block I, Unit 1 Symbolic AI
15

• Intelligent machines
• Search capable to locate the answer to a problem
  by sifting all possible answers and select the
  correct (or the best) one
• Symbols and rules can manipulate words (symbols)
  according to logical and linguistic rules
• Mathematical structure the implemented model
  must be a logical or mathematical structure of
  some kind.
• Randomness injection some degree of randomness
  into the orderly processes
• Neuron Networks simulating the structure found
  in the human brain

16
Block I, Unit 1 Symbolic AI
16

• Intelligent machines, two important keys
• Representation intelligent computer systems
  contain a model in some logical or mathematical
  form, of the problem being solved. These models
  are thus essentially symbolic, consisting of
  logical expressions
• Search computer systems can find intelligent
  answers to complex problems by searching among
  all possible answers for the best one. The
  process of search will be governed by rules.

17
Block I, Unit 1 Intelligence
17

• What is Intelligence ?
• The ability to comprehend, to understand and to
  profit from experience
• A general mental capability that involves the
  ability to reason, plan, solve problems, think
  abstractly, comprehend ideas and language and
  learn
• The ability of an individual to understand and
  cope with the environment
• The capacity to create constructively for the
  purpose of evolutionary gain

18
Block I, Unit 1 Intelligence
18

• Observations
• Observation 1 There is an obvious lack of
  agreement on what intelligence is, and thus of
  the exact goals of artificial intelligence.
• Observation 2 The only really clear and
  effective definitions of intelligence are in
  terms of a few examples of intelligent behavior
  perception, reasoning and action, in the case of
  Winston above decision making, problem solving
  and learning in Bellmans definition.
• Observation 3 The overwhelming focus is on
  human intelligence.

19
Block I, Turing work
19

• Alain Turing (1912, 1954) the father of AI
• Code breaker during the second world war
• Turing machine (invented on paper, 1936), it
  consists of
• a read/write head (or 'scanner') with a paper
  tape passing through it
• The tape is divided into squares, each square
  bearing a single symbol
• This tape is the machine's general purpose
  storage medium, serving both as the vehicle for
  input and output and as a working memory for
  storing the results of intermediate steps of the
  computation.

20
Block I, Turing work
20

• The machine can
• read (i.e. identify) the symbol currently under
  the head
• write a symbol on the square currently under the
  head (after first deleting the symbol already
  written there, if any)
• move the tape left one square
• move the tape right one square
• change state
• halt.
• Turing test

21
Block I, Unit 1 Cybernetics Vs. Symbolic AI
21

• Weak AI computer value is that it gives us a
  very powerful tool.
• Strong AI computer is not only a tool, rather,
  the appropriately programmed computer really is a
  mind

22
Block I, Unit 1 Computers
22

• The digital computers
• Formal systems
• Taken
• States/starting state
• Rules
• Automatic formal system one that works by itself
• Deterministic
• Non deterministic
• Heuristics experience based techniques for
  problem solving

23
Block I, Unit 1 Computers
23

• What computers can do?
• Models, for a natural system we have
• Simulation is a model that captures the
  functional connections between inputs and outputs
  of the system
• Replication is a model that captures the
  functional connections between inputs and outputs
  of the system and is based on processes that are
  same as, or similar to, those of the
  real-world-system
• Emulation is a model that captures the
  functional connections between inputs and outputs
  of the system and is based on processes that are
  same as, or similar to, those of the
  real-world-system and in the same materials as
  the natural system.

24
Block I, Unit 1 Computers
24

• Optimization problems
• Traveling Salesman Problem (TSP)
• For 5 cities, 120 possible combinations
• For 10 cities, 3.628.800 combinations
• For 15 cities, 1.307.674.368.000 combinations
• For 20 cities, 2.43 ? 1018
• Combinatorial explosion
• TSP is an NP hard problem. (there is no known
  algorithm for solving it in any realistic period
  of time, although such algorithm may exist)