Einingalíkön og hermun


1. Introduction to Agent-Based Modeling & Simulation




  Scientific Motivation

A large part of our universe remains elusive:
  - biological processes
  - ecosystems
  - human intelligence
  - social processes and commerce

  Agent-Based Modeling Among the newer methods used for addressing such "non-linear" problems - A combination of complex systems modeling and simulation will ensure the continued march of science in the above areas
  Practical Motivation Software simulations increasingly used by various scientific disciplines to study their phenomena of interest
Simulations used to be done by paper and pencil; now all are done in software
  Computer simulation Modern computers are powerful simulation platforms. They are underused. Some predictions: Computational power equaling that of the humand mind may be purchased for €1k in 2025, at the latest.








Main Concepts
  Agents/modules (Icel. einingar) & Interactions (Icel. samskipti)
  Interaction The behavior of one Agent affects the behavior of another Agent
  Agent Agent is a process with a behavior that transforms information from one format to another














Characteristics of Agents
  Omnipotence Can do anything
  Omniscience Knows everything
  Agent Neither omniscient nor omnipotent
  The prototypical agent is an animal - with a definite embodiment, which means that they are localized in "the world" (where "world" can have different meanings)
  They only see a small part of the world Have a limited view of the world
Their input (e.g. "perception") is limited
  Their abilities to act are limited and specialized Have a limited repertoire of actions the can inflict on the world
Their output (e.g. decision to act or action) is limited
  Their minds are limited, in one or more ways They have limited processing and memory capacity
  They may have other limitations which affect their functioning  
  The simplest ("dumb") agent acts alone That is, the simplest agent does not communicate or negotiate with other agents
  The smartest agents form "social networks"  











Models & Simulation
  Model A simplification of reality
Attempt to "get at the essence" of the phenomenon under study — stripping it down to its essentials
  Occam's razor "One should not increase, beyond what is necessary, the number of entities required to explain anything." [REF]
Einstein: "Everything should be made as simple as possible, but not simpler."
  Simulation vs. mathematical models Simulation is good for modeling phenomena that are too complex for mathematics; functions of all kinds, not just continuous functions, can be used













Agent-Based Approach
  No omniscience or omnipotence Exception: The experimenter, who typically may at any point interfere with anything in the simulation, and can (potentially) be said to "see everything" from a top-down view.
  Agents/modules/units Modules have input and output and some processing
  Input e.g. sensation and perception
  Output e.g. decisions
  State-less processing Processor with no memory, gets handed data along with instructions for anything it is supposed to do; produces output that contains results (plus side effects, if any)
  Local vs. centralized data storage Processors with local memory can afford simpler input and output
  Architecture Information Flow: Where is the information? Who controls how it flows?
Processing: Who does the processing? Where?
How models handle time: Continuous vs. Discrete (steps)














Related Technologies
  "Boids" - Craig Reynolds http://www.red3d.com/cwr/boids/   http://www.red3d.com/cwr/ibm.html
  SWARM, Santa Fe Institute http://www.santafe.edu/~mgd/lanl/framework.html
  Constructionist A.I. Thorisson et al. 2004 presented work specifically geared towards building A.I. systems
  Constructionist Design Methodology Derived from Constructionist A.I., in development by members of MINDMAKERS.ORG













Constructionist Design Methodology (CDM)
  Agent-based methodology Main CDM Link
  Provides step-by-step instructions for building modular systems  
  Blackboards Construct from early days of A.I., revived in CDM

- Old idea that is slowly filtering into the World Wide Web and business environments.
- Powerful for routing information where you don't know who might be interested in it.

  XML eXtensible Markup Language
  Message-based Message format from OpenAIR
  OpenAIR Data format and routing protocol
  Psyclone Software environment based on CDM implementing an OpenAIR server