L5: Creating Agent-Based Models

How can you verify an agent-based model?

by ensuring that a computational model faithfully implements its target conceptual model

What are the 3 steps to create an agent-based model?

• Designing your model

• Building your model

• Examining your model

What are the 2 parts to consider when designing a model?

• a conceptual/textual description

• a software-based implementation

Two major categories of ABMs

• Phenomena-based modeling

  • Model captures the referent pattern (e.g., segregation, spiral formation, oscillating populations of agents/species)

• Exploratory modeling

  • Create agents, define their behavior, and explore the patterns that emerge

What do research questions look like in the design of models?

• “How does a colony of ants forage for food?”

• No clear question, but want to model X and see what happens

What are the two approaches of Conceptualizing vs coding models?

• top-down

• bottom-up

In practice: A combination of approaches

What is the top-down approach of Conceptualizing vs coding models?

• Develop entire conceptual model first then implement it

• Need to have research question, design agents and their rules for behavior, elements of the situation/environment

• Refine and revise until it has enough detail to be coded

What is the bottom-up approach of Conceptualizing vs coding models?

Choose a domain of interest, start coding something relevant to the domain, adding in the conceptualizations, mechanisms, properties, entities along the way

What is the ABM design principle?

Start simple and build toward the question you want to answer.

• Start with the simplest set of agents and behaviors relevant to your topic

• Always keep your research question in mind (avoid unnecessary stuff)

How does verification work?

• Ensuring that a computational model faithfully implements its target conceptual model

• Simpler models are easier to verify, and scale up from

• Each version of the model should be linked to a research question

How do you Choose your research questions?

• Example for Wolf Sheep Model: “How do the population levels of two species change over time when they coexist in a shared habitat?”

• Make sure the modeling method fits with the questions (agent-based vs. equation-based)

  • Heterogeneous agent inhabiting space vs values following a function

• Key aspect is time (how X changes over time)

What are 6 aspects of Choosing your agents?

• Agent properties

• Environmental characteristics and stationary agents

• Agent behavior

• Time steps

• Parameters

• Measures

What is a Summary of wolf sheep simple model design?

• Driving Question: Under what conditions do two species sustain oscillating positive population levels in a limited geographic area when one species is a predator of the other and the second species consumes limited but regenerating resources from the environment?

• Agent Types: Sheep, Wolves, Grass

• Agent Properties: Energy, Location, Heading (wolf and sheep), Grass-amount (grass)

• Agent Behaviors: Move, Die, Reproduce (wolf and sheep), Eat-sheep (wolf only), Eat-grass (sheep only), Regrow (grass)

• Parameters: Number of Sheep, Number of Wolves, Move Cost, Energy Gain From Grass, Energy Gain From Sheep, Grass Regrowth Rate

• Time Step:

1. Sheep and Wolves Move

2. Sheep and Wolves Die

3. Sheep and Wolves Eat

4. Sheep and Wolves Reproduce

5. Grass Regrows

• Measures: Sheep Population versus Time, Wolf Population versus Time

What are 7 Key questions for top-down building your model?

1. What part of your phenomenon would you like to build a model of?

2. What are the principal types of agents involved in this phenomenon?

3. In what kind of environment do these agents operate? Are there environmental agents?

4. What properties do these agents have (describe by agent type)?

5. What actions (or behaviors) can these agents take (describe by agent type)?

6. How do these agents interact with this environment or each other?

7. If you had to define the phenomenon as discrete time steps, what events would occur in each time step, and in what order?

What are 3 aspects of Examining your model?

Multiple runs

• BehaviorSpace tool in Netlogo, Python code, or R control for netlogo

• Keep time steps constant but keep random seeds

Parameter sweeping and results collation

• Robustness and sensitivity analysis

Data analysis

• Statistical inference, visualization, etc.