11/04/2011

Emergent Computing

Agent-based modelling is a fairly new area of research in economics (economics is a bit behind the times compared to physics or biology here), and is something I'd like to look at (hopefully) during PhD study next year. Its fairly obviously related to cellular automata, in which can arise a phenomenon called "emergent computation" - decentralized information processing at a global level.  This kind of thing sounds very similar to the Austrian assertion that markets are essentially information-processors, and that 'the market' can create an allocation based on information that no agent individually possesses.  In other words, Austrian economists claimed that decentralized social systems should exhibit emergent computation - something that might be demonstrable with agent-based modelling.  This is obviously going on faith - I currently don't know anything about cellular automata.

People to read re this: 
  1. Melanie Mitchell at Sante Fe
  2. A New Kind of Science, Stephen Wolfram
  3. Cellular Automata, Andrew Ilachinski