Impacts of Diversity on Complex Systems

“Eye-­‐popping:  When  par1cles   collide  at  the  speed  of  light  in  a   vacuum  colder  than  -­‐271   Celsius,  they  put  on  a   spectacular  show  of  light,  heat   and  movement  that  form   complex  and  oBen  intricate   paCerns”     Source  of  figure  and  quote:   hCp://www.dailymail.co.uk/sciencetech/ ar1cle-­‐2021122/The-­‐Big-­‐Bang-­‐pictures-­‐ Scien1sts-­‐produce-­‐images-­‐par1cle-­‐ explosions-­‐similar-­‐greatest-­‐galac1c-­‐light-­‐ show.html#ixzz2fS8kCLrp

 

 

Save your time - order a paper!

Get your paper written from scratch within the tight deadline. Our service is a reliable solution to all your troubles. Place an order on any task and we will take care of it. You won’t have to worry about the quality and deadlines

Order Paper Now

What  phenomena  or  proper1es  emerge  from   having  diverse  types  that  interact?

 

 

Impacts  of  Diversity  on  Complex   Systems:  Synergies

•  Specializa1on:  compara1ve  advantage   •  Responsiveness   •  Compe11on   •  Superaddi1ve  tricks   •  Diversity  produc1on   •  Collec1ve  knowledge   •  Redundancy  (and  degeneracy)   •  Modularity  and  firewalling   •  CrosscuYng  cleavages

 

 

Specializa1on:  Compara1ve  Advantage

•  If  we  have  two  agents  that  each  do  something   beCer  than  the  other,  than  it  becomes   advantageous  to  specialize  and  divide  labor

•  Thus,  compara1ve  advantage  is  a  driver  of   specializa1on,  and  favors  the  evolu1on  of   trade  or  coopera2on  among  specialized   agents

 

 

Specializa1on:  Learning  by  Doing

•  The  faster  the  rate  of  learning,  the  larger  the   compara1ve  advantage

•  We  can  think  of  learning  in  two  ways:

–  Individual  agents  learn  over  the  course  of  their  life1mes,   i.e.,  they  improve  at  tasks  the  more  they  do  them

–  Evolu1on  “learns”  through  the  process  of  explora1on  and   exploita1on,  i.e.,  adapta1on  is  a  form  of  learning  by   gene1c  informa1on  systems

 

 

Responsiveness

•  The  ability  of  a  system  to  respond  to   disturbances

•  The  greater  the  diversity  of  possible  responses   (size  of  the  set  of  the  possible),  the  more   disturbances  a  system  can  absorb

 

 

Compe11on   •  Compe11on  drives  the  evolu1on  of  diversity  by  (1)   favoring  variants  that  “do  beCer”  and  (2)  selec1ng  for   agents  to  be  more  different

•  Compe11on  can  be  thought  of  as  conflict  and  the   evolu1onary  outcomes  “do  beCer”  and  “be  more   different”  as  forms  of  conflict  resolu2on

•  In  the  evolu1on  of  biological  diversity,  there  is  a   constant  dynamic  between  compe11on  (conflict)  and   selec1on  for  compe1tors  to  adapt  (innovate)  and   diverge  (conflict  resolu1on)

 

 

Superaddi1ve  Tricks

•  A  func1on  f  is  said  to  be  superaddi1ve  if:      f  (x  +  y)  >  f  (x)  +  f  (y)

•  Superaddi1vity  allows  lots  of  room  for   innova1on,  and  makes  innova1on  cheap

 

 

Diversity  Produc1on

 

Diversity  Begets  Diversity

 

 

Collec1ve  Knowledge

•  The  ability  of  a  collec1ve  to  come  to  an  accurate   forecast  or  to  decide  on  an  appropriate  ac1on

•  The  theory:  if  group  members  individually  make   different  par11ons  and  predic1ons  about  the   world,  then  collec1vely  groups  should  make   more  accurate  predic1ons  of  the  world  and   therefore  take  more  appropriate  ac1ons

 

 

Redundancy  and  Degeneracy

•  If  a  system  contains   redundant  parts,  then  it   will  be  robust  to  the   failure  of  any  one  of   those  parts

•  Predic1on:  more   redundant  ecosystems   are  more  robust   ecosystems

 

 

Modularity  and  Firewalling

“The  figure  shows  modules  of  a  puzzle  which  can   form  different  shapes  when  they  are  placed  at   different  places.  The  modules  can  be  moved  freely   without  affec1ng  the  func1onality  of  other  modules   but  it  changes  the  system’s  shape(func1onality).”

hCp://www.defit.org/2012/06/modularity.html  Source  of  figure  and  quote:  IT  Defini1ons

 

 

Group  C

CrosscuYng  Cleavages  –  group   membership  context

Group  A   Group  B

Some  members  of  each  group   have  membership  in  all  other   groups

 

 

Popula1on  C

CrosscuYng  Cleavages  –  gene1c   cohesion  context

Popula1on  B  Popula1on  A

Some  members  of  each  popula1on   share  genes  from  all  other  popula1ons