Some Questions

1
Some Questions

• At what level(s) do we define an organism?
  Does it matter? How about systems in general?
• If it is true that there is a universal law of
  vivisystems -- no matter how deeply we
  investigate a subunit of a systems, we cannot
  infer the whole -- what does this imply for
  educational research and educational improvement?
• How important is it that many phenomena in the
  world can be described by the S-curve?

2
Grand Aspirations

• Kelly looking for unifying principles off all
  large vivisystems calling them the laws of god
• Casti trying to develop the science of
  surprise
• Waldrop telling the story of researchers hoping
  to find commonality between physics and economics
  (and other fields)

3
Clockwork Logic vs. Bio-logic

• Kelly Chapter 1
• Bio-logic being transferred mechanical systems
• Self-replication
• Self-governance
• Limited self-repair
• Mild evolution
• Partial learning
• Clockwork logic being transferred to biology
• I.e., bioengineering

4
Characteristics of Swarm systems

• Also called networks, complex adaptive systems,
  vivisystems, collective systems, (dynamical
  systems?)
• The absence of imposed centralized control
• The autonomous nature of subunits
• The high connectivity between the subunits
• The webby nonlinear causality of peers and
  influencing peers

5
Examples of Swarm Systems

• Bee hives
• Conference attending playing Pong and flying
  aircraft with green and red lights
• The mind recreating memory
• N-body problem?
• Voting / societal preference aggregation?

6
Advantages of Swarm Systems

• Adaptable
• Evolvable
• Resilient
• Boundless
• Produce Novelty

7
Disadvantages of Swarm Systems

• Non-optimal
• Non-controllable
• Non-predictable
• Non-understandable
• Non-immediate

8
Attractors

• However, Casti says that some of these systems
  may have attractors
• 3 types
• Fixed point
• Periodic (limit cycle)
• Strange attractors (unstable periodic patterns)
• Domain of attraction

9
Features of Dynamical Systems

• From Casti p. 35
• Small changes in system can lead to a large
  divergence in outcomes
• Randomness does not have to come from
  uncertainty. It can come from deterministic
  rules.
• Unstable equilibria, or instability of
  itineraries

10
An Empirical Observation

• Dynamics of many things in the world follow the
  pattern below (Marchetti)

Ffraction of final amount
Amount of X
F/(1-F)
Time
Time
11
Castis Attempt at a Science of Surprise

• What are the implications of all this complex
  systems stuff?
• Hard to know exactly, but Casti takes a crack at
  identifying common surprise generators
• Logical tangles (e.g., Arrows Impossibility
  Theorem)
• Instability (e.g., agglomeration of high-tech
  firms)
• Uncomputability (e.g., wave motion)
• Irreducibility (e.g, N-body problem)
• Emergence (e.g., Kauffman nets)

12
Interesting Implications (for me)

• Caused Brian Arthur a lot of anguish.
• He realized that many more economic phenomena are
  better thought of as complex adaptive systems
  than economists were willing to admit. (Chap 1,
  Waldrop.)
• Organisms can be defined at multiple levels
• Cannot make inferences about the whole only from
  deep investigation of individual parts
• Where supreme control is needed, use clockware
  where supreme adaptability is needed, use
  swarmware
• Sensitivity to initial conditions even the most
  deterministic of systems can display wildly
  divergent behavior with small changes (e.g.,
  Circle-10 system)