From Natural to Artificial Systems Models of Competition and - - PowerPoint PPT Presentation
From Natural to Artificial Systems Models of Competition and Models of Competition and Cooperation Cooperation By Rob Cranston, Walter Proseilo, Chau Trinh & Owen Pang Table of Contents K Introduction K Modeling a Society of Mobile
By Rob Cranston, Walter Proseilo, Chau Trinh & Owen Pang
K Introduction K Modeling a Society of Mobile
K Transmitting Culture K Deciding Whether to Interact K Choosing How to Behave K Summary
O An agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through effectors. (from Intelligent Agents by
O Powerful Multi-Use Tool. O Thousands of built in
functions.
O Easy to use programming
tool.
O Used for all simulations in
this presentation.
ORule based language – good for simulations OVery strong pattern matching ORules for our simulations rely on this. The
pattern matching is used to determine which rule is carried out on the agent
O “Simulating Society” by
Gaylord & D’Andria
O Simulations involving
groups of agents
O Builds on others work and
uses Mathematica as the tool for the simulations
O All simulations in our
presentation are from this book
ODecentralized ODiscrete ODynamic
OSpace is represented in 2-D OEach cell is defined as a state OThe system evolves over time OCells updated using rules
OSquare n x n lattice OPopulation of density - p OThe system evolves time steps - t
OAn empty site has a value of 0 OA site occupied by an individual has a value
which is a list
Note: it is useful to focus on the lattice sites rather than on the individuals.
OTime step is executed in two or more
consecutive partial-steps
OIn each partial-step, a set of rules is applied to
each site in the lattice
OOne agent per cell ONeighborhood ODirection OWalk rules for updating
a lattice site have the form: walk[site, N, E, S, W, NE, SE, SW, NW, Nn, Ee, Ss, Ww]
Ww Ee Nn Ss SW NW NE W S N E SE
Each lattice occupied by an agent becomes empty unless: Cell remains occupied by the agent, who chooses a random direction to face
Scenario #1 Scenario #2
OPerson to Person OPerson to Group
OThe system evolves over t time steps, starting
with the initial lattice configuration and society
Step 1 Step 2 Step 3 Step 498 Step 499
OConsists of a Meme
list of Features and Traits
OA = {3, 2, 1, 7, 5} ON = {4, 8, 1, 2, 5}
A N
The System
O A = {3, 2, 1, 7, 5} O N = {4, 8, 1, 2, 5}
Cultural Exchange
O A = {3, x, 1, 7, 5} O N = {4, 8, 1, 2, 5}
Where x is a randomly chosen integer between 2 and 8. A N
OIncorporating mobility OIncorporating bilateral
cultural exchange
OSocial Status and
Role Models
Bill Gates
OGood behavior versus bad behavior
OPayoffs resulted from interaction OBenefit if positive payoff OCost if negative payoff
OSquare n by n lattice
OEmpty site has 0 OGood & Bad guys OSite occupied by an individual has a list
I = {a, b, c, d, e}
I
OMemory Checking ORefuse or Accept Interaction OUpdate List
Graph of Good Guy vs. Bad Guy
Graph of Good Guy vs. Bad Guy
Graph of Good Guy vs. Bad Guy
Signals
“I suggest you deactivate your emotion chip for now.” Patrick Stewart in Star Trek: First Contact (1996)
http://www.geocities.com/Area51/Vault/126/
Graphs of Good Guys and Bad Guys
Introduction
Too many variations of UNIX Setting a Standard UNIX International Inc. (UII) Open Software Foundation (OSF) Two types of Companies
Uses Landscape Theory
size: si propensity: pij configuration: X distance: dij frustration: Fi(X) energy: E(X)
Assumptions
Cooperation Competition
O Additional parameters α and β used to indicate close
rivals
O Nash Equilibrium
Results: Only two configurations that were also Nash
Equilibriums
Alliance 1 Alliance 2
Sun DEC AT&T HP Prime Apollo IBM Intergraph SGI
Configuration A Alliance 1 Alliance 2
Sun AT&T DEC Prime HP IBM Apollo Intergraph SGI
Configuration B
S p e c i a l i s t G e n e r a l i s t
Being good vs. being bad Adaptation Introspection
Based on the Behavioral History of the Other
Individual
Reciprocity
Pollyanna Sociopath Nice retaliator Mean retaliator
Square n by n lattice
Empty site has 0 Site occupied by an individual has a list
I = {a, b, c, d, e}
I
Deciding Interacting Moving
Introspective
model
Satiation
By Rob Cranston, Walter Proseilo, Chau Trinh & Owen Pang
OSummary OQuestions OWebnotes:
http://www.cpsc.ucalgary.ca/~pango/533/
March 27th Revision 4