Games and Agents

Session I - Introduction to Agent Based Modeling

Ali Seyhun Saral (Uni. Bologna)

Institute of Psychology at Leiden University.

13 June 2022

A flock of birds

A flock of birds

Simple rules lead to seemingly sophisticated behavior (Reynolds 1987)

  • Align yourself with the neighbors in terms of velocity and direction.
  • Separating yourself enough from the others to avoid collusion.
  • Try to center yourself to your neighbors.

Simple rules -> Complex Behavior

What is a scientific model?

  • A simplified representation of some part of the reality that is useful for a scientific purpose.
    • Help us to understand the reality it represents.
    • Manipulate it to improve causal understanding.
    • Make predictions out of it.

“LEGO House” by georgivar is licensed under CC BY-SA 2.0.

Modelling complex systems

  • The complexity of scientific models was limited by its mathematical tractability.

  • Computer simlations relaxed relaxed this limitation.

  • ABM’s are less simplified in a specific way:

    • They describe individual components of a system
    • Individuals and components are autonomous entities

Agent-based models

  • Computational model for simulation the interactions of autonomous objects (agents).
  • Agents: people, animals, institutions, particles, etc.
  • Bottom-up approach
  • Define the behavior of agents and their interactions.

What problems are addressed by ABM?

(See Railsback and Grimm 2019. For more, p.11)

Why ABM?

  • Solve problems that traditional models and methods are too simple for.
  • Support/refute theory based on additional features.
  • Explore potential distribution of outcomes.
  • Adresses problems related to emergence.

Emergence

  • Describes a relationship between a low-level system (micro) and a high-level (macro) system.

  • Macro-level phenomena can only be derived by studying micro-level phenomena.

  • Simple interactions between a system’s elements lead to unexpected global behavior (Epstein 1999)

Examples:

  • The human body
  • The human brain
  • Pandemics
  • Ecosystems
  • Global Production chain
  • Mexican waves

Features of ABM

(Romanowska, Wren, and Crabtree 2021)

  • There are no strict rules on how ABM should be built.

Some features:

  • Emergent
  • Heterogeneity
  • Temporal
  • Spatial
  • Learning/adaptation

Complex Adaptive Systems and ABM

  • When we use agent-based models, we useally study a compleyx adaptive system.
  • Dynamic of systems composed if interacting elements in different levels.

Two historical inspiration

  • Evolution

  • Emergence

Evolution: Universal Constructor

  • One robot buiding another
    • Without a computer
    • Design a machine which’s complexity increase under natural selection.
    • The concept of celular automata

John von Neumann (1903-1957)

Game of Life (1970)

  • A cellular automaton
  • Square cells with two states: alive and dead.
  • 8 neigbors:
    • if <2 alive neighbors: dies
    • if 2-3 alive neighbors: lives
    • if > 3 alive neighbors: dies
    • if 3 alive neighbors: lives

Game of Life (1970)

Full Video: Youtube

Axelrod Tournaments (1980)

  • Iterated Prisoners Dilemma Game: “Cooperate” “Defect”
  • Colleagues were invited to submit their strategies
  • Strategies played against each other repeatedly
  • 15 strategies initially.
  • The winner was the TIT FOR TAT strategy.

Building Agent-Based Models

  • Modeling (Design, description etc.)
  • Programming

What software to use

https://en.wikipedia.org/wiki/Comparison_of_agent-based_modeling_software

  • Netlogo

  • Python

  • Mesa (Python)

  • AgentPy (Python)

  • R

Why Python?

  • The skills are transferable
  • The language is easy to learn
  • You get to understand underlying concepts of ABM
  • Potential to use in other fields with packages
  • Easier to connect wit data
  • Potential to combine other technologies
    • Reinfocement learning
    • Regression analysis

This workshop

  • Hands-on introduction to agent based models
  • We will use Python
  • We will build models from scratch
  • We will use other packages like Axelrod, AgentPy
  • Focus more on Game Theory

Workshop Program

13 June 2022, Monday

  • Python Basics

14 June 2022, Tuesday

  • Python: Introduction to Object-Oriented Programming in Python
  • Creating an Agent
  • Interacting agents
  • Implementing the Game
  • Axelrod Tournaments

Workshop Program- cont

15 June 2022, Wednesday

  • Creating a Population
  • Evolution of strategies
  • Python: Numpy, Pandas, Matplotlib
  • Sensitivity Analysis

16 June 2022, Thursday

  • Grid Games
  • Seggregation Model
  • Networks
  • Advanced Topics Discussion

Preperation

  • Jupyter Notebook

References

Epstein, Joshua M. 1999. “Agent-Based Computational Models and Generative Social Science.” Complexity 4 (5): 41–60.
Geanakoplos, John, Robert Axtell, J Doyne Farmer, Peter Howitt, Benjamin Conlee, Jonathan Goldstein, Matthew Hendrey, Nathan M Palmer, and Chun-Yi Yang. 2012. “Getting at Systemic Risk via an Agent-Based Model of the Housing Market.” American Economic Review 102 (3): 53–58.
Gilbert, Nigel, John C Hawksworth, and Paul A Swinney. 2009. “An Agent-Based Model of the English Housing Market.” In AAAI Spring Symposium: Technosocial Predictive Analytics, 30–35.
Hinch, Robert, William JM Probert, Anel Nurtay, Michelle Kendall, Chris Wymant, Matthew Hall, Katrina Lythgoe, et al. 2021. “OpenABM-Covid19—an Agent-Based Model for Non-Pharmaceutical Interventions Against COVID-19 Including Contact Tracing.” PLoS Computational Biology 17 (7): e1009146.
Huth, Andreas, Martin Drechsler, and Peter Köhler. 2004. “Multicriteria Evaluation of Simulated Logging Scenarios in a Tropical Rain Forest.” Journal of Environmental Management 71 (4): 321–33.
LeBaron, Blake. 2001. “Empirical Regularities from Interacting Long-and Short-Memory Investors in an Agent-Based Stock Market.” Ieee Transactions on Evolutionary Computation 5 (5): 442–55.
Railsback, Steven F, and Volker Grimm. 2019. Agent-Based and Individual-Based Modeling: A Practical Introduction. Princeton university press.
Reynolds, Craig W. 1987. “Flocks, Herds and Schools: A Distributed Behavioral Model.” In Proceedings of the 14th Annual Conference on Computer Graphics and Interactive Techniques, 25–34.
Romanowska, Iza, Colin D Wren, and Stefani A Crabtree. 2021. Agent-Based Modeling for Archaeology: Simulating the Complexity of Societies. SFI Press.