Complexity projects

  • Game of Life
  • Social distancing
  • Crowd simulation
  • Building evacuation
  • SIR (epidemiology)

A complex system has

  • __A__daptation and learning
  • __C__onnectedness
  • __I__nterdependence
  • __D__iversity

Wolfram behaviours

  • Stables
  • Periodic
  • Chaotic
  • Complex (high info content)

Misc

  • Exploration versus exploitation
  • Highest peak
  • On a dancing landscape you can never stop exploring
  • Black swan

Emergence

  • Complexity is an emergent property
  • Stasis encourages exploration and vice versa
  • Slime mold breaking symmetry
  • Bottom-up and top-down emergence
  • Power law distribution
  • Long tail
  • Weekly emerging, strongly emerging (never figure out)
  • Preferential attachment model

Agent-based models

  • Fires in crowded buildings
  • Epidemics
  • Netlogo
  • Good science and agent based models must simplify (abstract)

Feedback and externality

Feedback is affecting the same action.

  • Positive feedback yields tipping points
  • Negative feedback yields stability
  • Feedback and externality produce path dependence

The tail wags the dog

  • With complex systems the mean doesn’t matter: it’s the tails
  • Segregation is emergent
  • Tipping is one way only, segregation is stable

Social randomness – coin flipping

Humans underestimate how long runs of the same side can occur randomly. If a person is playing the part of the coin and selects three heads, the fourth flip won’t be a 50/50 chance of heads or tails – whereas the result of a truly random process would not be affected by the previous flips.

  • Coin flip return times
  • Don’t allow system to build themselves without adding extra checks to limit large events
  • Central limit theorem