Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
A riddle wrapped in a mystery inside a rich, flaky crust

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



A-life Society

Creating a genetically evolving artifical society of agents
(+1, -1)
  [vote for,

I once read an artical in the Atlantic, (http://www.theatlantic.com/) about some research into A-life Societies in which individual motive was being studied on a Societal scale. The simulations always ended up with some drastically oversimplifed motive for each agent, like "all agents are 50% tolerant of any other race". So I decided to make my own simulation with individual tolerances, likes and dislikes so that if the simulation outputted some descernable outcome it would at least seem reasonable. The question I've got though is what distribution do I use to randomize the population? Bell Curve? Just a random() assignment? Some other model? I can see why researchers don't incorporate individual preferences since it's just a shot in the dark to determine those preferences.

If that weren't hard enough, my simulation incorporates an preference evolution for an agent. Each agent's preference changes randomly and by a random amount in a random direction (up or down). This seems equally distasteful. I need a more realistic method to alter preferences but I also don't want to incorporate some psychoanalysis of other agents to determine other agent's preferences.

It seems this will end up as only another of my half-baked ideas.

sobokhan, Apr 24 2002

Please log in.
If you're not logged in, you can see what this page looks like, but you will not be able to add anything.
Short name, e.g., Bob's Coffee
Destination URL. E.g., https://www.coffee.com/
Description (displayed with the short name and URL.)

       A good way to start might be to use a statistical snapshot of an existing society.   

       As far as variations in a particular attribute (like tolerance) for a particular agent, well, they shouldn't be completely random (unless all your agents are insane). You need some external stimulus which might modify the attribute (possibly the other agents and their behaviour).
phoenix, Apr 24 2002

       Do you trust microeconomic principles to influence the ultimate behavior of your agents over time and change? Your decision trees become random only by obsolescence or scarcity -- not a very evolutionary medium in which to culture your society.   

       The central theorum is that individuals change their behavior to select an optimum basket of goods and services within a preference set that is both personal and collaborative. The only way to alter preferences within a universe of possible choices is to introduce randomness via scarcity or via obsolescence.
reensure, Apr 25 2002


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle