Aucun titre de diapositive

1

Knowledge Management and Management
LearningComplexity theory
Walter Baets, PhD, HDR Associate Dean for
Research MBA Director Professor Complexity ,
Knowledge and Innovation Euromed Marseille
Ecole de Management Erna Baets Oldenboom,
MA Professor Leadership, Sustainable Performance,
and Mind/Body Medicine
2
(No Transcript)
3
Flatland Edwin Abbott, 1884 A. Square meets
the third dimension
4
Taylors view on the brain
The computer attempt to automate human thinking
Manipulating symbols Modeling the
brain Represent the world
Simulate interaction of neurons Intelligence
problem solving Intelligence learning 0-1
Logic and mathematics Approximations,
statistics Rationalist, reductionist Idealized,
holistic Became the way of building computers
Became the way of looking at minds
5
Sometimes small differences in the
initial conditions generate very large
differences in the final phenomena. A slight
error in the former could produce a tremendous
error in the latter. Prediction becomes
impossible we have accidental phenomena.
Poincaré in 1903
6
Sensitivity to initial conditions (Lorenz)
Xn1 a Xn (1 - Xn)
0.294 1.4 0.3 0.7
7
Cobweb Diagrams (Attractors/Period Doubling)
Xn1 ? Xn (1 - Xn) (stepfunction) dX
/ dt ? X (1 - X) (continuous function)

• On the diagrams one gets
• Parabolic curve
• Diagonal line Xn1 Xn
• Line connecting iterations

8
Lorenz curve (Butterfly effect)
Lorenz (1964) was finally able to materialize
Poincarés claim Lorenz weather forecasting
model dX / dt B ( Y - X ) dY / dt -
XZ rX - Y dZ / dt XY - bZ
9
Hénon Attractor
X n1 1 - a X 2 n Y n Y n1 b X n
Again, different attractors are shown Other
examples Pendulum of Poincaré, Horse Shoe
10
Why can chaos not be avoided ?

• Social systems are always dynamic and
• non-linear
• Measurement can never be correct
• Management is always a discontinuous
• approximation of a continuous
• phenomenon

11
Ilya Prigogine

• Non-linear dynamic models (initial state,
• period doubling,.)
• Irreversibility of time principle
• The constructive role of time
• Behavior far away from equilibrium (entropy)
• A complex system chaos order
• Knowledge is built from the
• bottom up

12
Entropy
Measure for the amount of disorder When entropy
is 0, no further information is
necessary (interpretation is that no information
is missing There is a maximum entropy in each
system (in the bifurcation diagram, this is
4) Connection between statistical mechanics and
chaos is applying entropy to a chaotic system in
order to compare with an associated statistical
system
13
Francesco Varela

• Self-creation and self-organization of systems
  and structures (autopoièse)
• Organization as a neural network
• The embodied mind
• Enacted cognition
• Subject-object division is clearly artificial
• How do artificial networks operate (Holland)
• Morphic fields and morphic resonance (Sheldrake)

14
Chris Langton

• Artificial life research
• Genetic programming/algorithms
• Self-organization (the bee colony)
• Interacting (negotiating) agents

15
Conways game of life

• One of the earlier artificial life simulations
• Simulates behavior of single cells
• Rules
• Any live cell with fewer than two neighbors dies
  of loneliness
• Any live cell with more than three neighbors dies
  of crowding
• Any dead cell with exactly three neighbors comes
  to life
• Any cell with two or three neighbors lives,
  unchanged to the
• next generation
• Plife.exe (windows)

16
John Holland

• Father of genetic programming
• Agent-based systems (network)
• Individuals have limited characteristics
• Individuals optimize their goals
• Limited interaction (communication) rules

17
Law of increasing returns (Brian Arthur)

• Characteristics of the information economy
• (a non-linear dynamic system)
• Phenomenon of increasing returns
• Positive feed-back
• No equilibrium
• Quantum structure of business
• (WB)

18
Summary (until now)

• Non - linearity
• Dynamic behavior
• Dependence on initial conditions
• Period doubling
• Existence of attractors
• Determinism
• Emergence at the edge of chaos