Exploring Complexity with Agent-Based Models

Daniel Vartanian

University of São Paulo

August 15, 2024

Hi There! 👋

1. Introduction [2.5m]
2. Key Concepts of Complex Systems [15m]
3. Agent-Based Models [10m]
4. Real-World Applications [5m]
5. Final Remarks [2.5m]

This presentation invites you to explore the fascinating world of complex systems, generative science, and agent-based models. Together, we’ll uncover the key principles and tools that bring these systems to life, revealing how they can transform our understanding of socio-ecological dynamics.

We’ll explore the following topics:

1. Key Concepts of Complex Systems
2. Agent-Based Models
3. Real-World Applications
4. Final Remarks

Key Concepts of Complex Systems

Complex versus Complicated

Complex versus Complicated

Complex versus Complicated

A look at the internals of a biological clock (a timing system).

A mess! A bunch of coupled oscillators

“[…] involve great numbers of parts undergoing a kaleidoscopic array of simultaneous interactions.” (Holland, 1992)

If you look just its parts, you can’t understand the whole system.

It’s an emergent phenomenon.

It can be thought as an “aggregate agent” ― aggregate behavior of component agents generates behavior of the aggregate agent. (Holland, 2012)

Boundaries and signals (Holland, 2012)

It’s adaptive, robust, produces extreme events, and is self-organized.

Emergence

Stable macroscopic patterns arising from local interaction of agents (Epstein, 1999).

When the aggregate exhibits properties not attained by summation (Holland, 2014).

The behavior of each part don’t explain how they behave collectively.

Levels of description (Nicolis & Nicolis, 2012).

In mathematical terms, the interactions of interest are non-linear (Holland, 2014).

Emergence isn’t magic (Wilson, 2004).

You are dealing with an emergence phenomenon when there is no need to look under the hood (Krakauer, 2023).

Emergence

A aggregate behavior emerges from the interactions of the parts (CAS) (Holland, 2012).

Chaos

Deterministic nonperiodic flow (Lorenz, 1963).

Systems in which this is the case are said to be sensitively dependent on initial conditions (Lorenz, 2008).

Does the Flap of a Butterfly’s Wings in Brazil Set off a Tornado in Texas? (Lorenz, 2008)

The behavior of some simple, deterministic systems can be impossible, even in principle, to predict in the long term, due to sensitive dependence on initial conditions (Mitchell, 2009).

Order in chaos: You can’t predict how any individual state will evolve, but you can say how a collection of states evolves (Muller, 2019).

Seemingly random behavior can emerge from deterministic systems, with no external source of randomness (Mitchell, 2009).

See: Lorenz system.

Pseudonoise

It only looks random (Lorenz, 2008).

Chaoplexologists such as Wolfram assume that much of the noise that seems to pervade nature is actually pseudonoise, the result of some underlying, deterministic algorithm (Horgan, 2004).

Micromotives and macrobehavior (Schelling, 2006).

Generative Science

Simulation is a third
way of doing science.
(Axelrod, 1997, p. 24)

Like deduction, it starts with a set of explicit assumptions. But unlike deduction, it does not prove theorems. Instead, a simulation generates data that can be analyzed inductively.

Unlike typical induction, however, the simulated data comes from a rigorously specified set of rules rather than direct measurement of the real world.

While induction can be used to find patterns in data, and deduction can be used to find consequences of assumptions, simulation modeling can be used as an aid intuition.

Agent-Based Models

The main objective of this section is to introduce the concept of agent-based models (ABMs) and other important concepts related to modeling in general.

Agent-Based Models

Agent-based models (ABMs) are computational models with the purpose of simulating the behavior of agents and their interactions, allowing us to study emergent phenomena.

Instead of describing a system only with variables representing the state of the whole system (a global approach/top-down), we model its individual agents (a local approach/bottom-up) (Railsback & Grimm, 2019).

Agents often represent people or other animals, but agents can also represent anything from biological cells to economic firms to political municipalities (Smaldino, 2023).

The Modeling Cycle

We have to force ourselves to simplify as much as we can, or even more. The modeling cycle must be started with the most simple model possible, because we want to develop understanding gradually, while iterating through the cycle. A common mistake of begin- ners is to throw too much into the first model version—usually arguing that all these factors are well known and can’t possibly be ignored. The modeling expert’s answer to this is, yes, you might be right, but—let us focus on the absolute minimum number of factors first. Put all the other elements that you think might need to be in the model on your “wish list” and check their importance later (Railsback & Grimm, 2019, p. 8).

Conceptual Models

Key Components of ABMs

Agents

Environment

Interaction

When to Use ABMs?

• Medium numbers
• Heterogeneity
• Time
• Rich environments
• Adaptation
• Complex but local interactions

ABM Frameworks

• NetLogo
• StarLogo
• Swarm
• GAMA
• Mesa (Python)

See also: Comparison of agent-based modeling software.

Real-World Applications

Historical Ecology

Agent-based model of the ancient Maya social-ecological system (applied archaeology) (NetLogo/Scala-Java).

Systems Biology

Multi-scale simulation of LNCaP prostate cancer cell line and combinations of drugs (PhysiBoSS/C++).

Urban Planning

Agent-based simulation tool that supports simulation of the evacuation of a city’s population at fine temporal and geographical scales (GAMA/Java).

ABM + AI: Agent hospital

A simulacrum of hospital with evolvable medical agents.

Final Remarks

How to Learn More?

Here are some resources to help you get started:

🎓 Courses

Introduction to complexity (Santa Fe Institute)

Introduction to agent-based models (Santa Fe Institute)

Introdução à ciência da computação com python - Parte 1 (USP-IME)

Introdução à ciência da computação com python - Parte 2 (USP-IME)

Data science specialization (Johns Hopkins)

How to Learn More?

Here are some resources to help you get started:

📝 Articles

Foundational papers in complexity science

Epstein, J. M. (1999). Agent-based computational models and generative social science. Complexity, 4(5), 41–60. https://doi.org/10.1002/(SICI)1099-0526(199905/06)4:5<41::AID-CPLX9>3.0.CO;2-F

Grimm, V., Railsback, S. F., Vincenot, C. E., Berger, U., Gallagher, C., DeAngelis, D. L., Edmonds, B., Ge, J., Giske, J., Groeneveld, J., Johnston, A. S. A., Milles, A., Nabe-Nielsen, J., Polhill, J. G., Radchuk, V., Rohwäder, M.-S., Stillman, R. A., Thiele, J. C., & Ayllón, D. (2020). The ODD protocol for describing agent-based and other simulation models: a second update to improve clarity, replication, and structural realism. Journal of Artificial Societies and Social Simulation, 23(2), 7. https://doi.org/10.18564/jasss.4259

How to Learn More?

Here are some resources to help you get started:

🎥 Videos

Krakauer, D. (2023, February 17). What is complexity? [YouTube video]. Santa Fe Institute. https://www.youtube.com/watch?v=JR93X7xK05o

📙 Books

Mitchell, M. (2009). Complexity: A guided tour. Oxford University Press.

Wilensky, U., & Rand, W. (2015). An introduction to agent-based modeling: modeling natural, social, and engineered complex systems with NetLogo. The MIT Press.

Railsback, S. F., & Grimm, V. (2019). Agent-based and individual-based modeling: a practical introduction (2. ed.). Princeton University Press.

Closing Remarks

This presentation was created using the Quarto Publishing System. Code and materials are available on GitHub.

Several of these slides benefited from significant contributions from Bill Rand, Camilo Rodrigues Neto, and others.

References

In accordance with the American Psychological Association (APA) Style, 7th edition.

Axelrod, R. (1997). Advancing the art of simulation in the social sciences. In R. Conte, R. Hegselmann, & P. Terna (Eds.), Simulating Social Phenomena (Vol. 456, pp. 21–40). Springer. https://doi.org/10.1007/978-3-662-03366-1_2

Carvalho, A. M. de, Garcia, L. M. T., Lourenço, B. H., Verly Junior, E., Carioca, A. A. F., Jacob, M. C. M., Gomes, S. M., & Sarti, F. M. (2024). Exploring the nexus between food systems and the global syndemic among children under five years of age through the complex systems approach. International Journal of Environmental Research and Public Health, 21(7, 7), 893. https://doi.org/10.3390/ijerph21070893

Daudé, E., Chapuis, K., Taillandier, P., Tranouez, P., Caron, C., Drogoul, A., Gaudou, B., Rey-Coyrehourq, S., Saval, A., & Zucker, J.-D. (2019, May). ESCAPE: Exploring by simulation cities awareness on population evacuation. ISCRAM. https://hal.science/hal-02130387

Epstein, J. M. (1999). Agent-based computational models and generative social science. Complexity, 4(5), 41–60. https://doi.org/10.1002/(SICI)1099-0526(199905/06)4:5<41::AID-CPLX9>3.0.CO;2-F

Heckbert, S., Isendahl, C., Gunn, J. D., Brewer, S., Scarborough, V. L., Chase, A. F., Chase, D. Z., Costanza, R., Dunning, N. P., Beach, T., Luzzadder-Beach, S., Lentz, D. L., & Sinclair, P. (2019). Growing the ancient maya social-ecological system from the bottom up. In C. Isendahl & D. Stump (Eds.), The Oxford handbook of historical ecology and applied archaeology (pp. 302–322). Oxford University Press. https://doi.org/10.1093/oxfordhb/9780199672691.013.30

Holland, J. H. (1992). Complex adaptive systems. Daedalus, 121(1), 17–30. https://www.jstor.org/stable/20025416

Holland, J. H. (2012). Signals and boundaries: Building blocks for complex adaptive systems. MIT Press. https://doi.org/10.7551/mitpress/9412.001.0001

Holland, J. H. (2014). Complexity: A very short introduction. Oxford University Press.

Horgan, J. (2004). The end of science revisited. Computer, 37(1), 37–43. Computer. https://doi.org/10.1109/MC.2004.1260723

Krakauer, D. (2023, February 17). What is complexity? [Video recording]. Santa Fe Institute. https://www.youtube.com/watch?v=JR93X7xK05o

Lewin, R. (1993). Complexity: Life at the edge of chaos. Collier Books.

Li, J., Wang, S., Zhang, M., Li, W., Lai, Y., Kang, X., Ma, W., & Liu, Y. (2024, May 5). Agent hospital: A simulacrum of hospital with evolvable medical agents. https://doi.org/10.48550/arXiv.2405.02957

Lorenz, E. N. (1963). Deterministic nonperiodic flow. Journal of the Atmospheric Sciences, 20(2), 130–141. https://doi.org/10.1175/1520-0469(1963)020<0130:DNF>2.0.CO;2

Lorenz, E. N. (2008). The essence of chaos (Publicado originalmente em 1995.). UCL Press.

Mitchell, M. (2009). Complexity: A guided tour (OCLC: ocn216938473). Oxford University Press.

Muller, D. (2019, December 6). Chaos: The science of the butterfly effect [Video recording]. Veritasium. https://www.youtube.com/watch?v=fDek6cYijxI

Nicolis, G., & Nicolis, C. (2012). Foundations of complex systems: Emergence, information and prediction (2nd ed.). World Scientific.

Pérez-Villa, A., Echeverría-Garcés, G., Ramos-Medina, M. J., Prathap, L., Martínez-López, M., Ramírez-Sánchez, D., García-Cárdenas, J. M., Armendáriz-Castillo, I., Guerrero, S., Paz, C., & López-Cortés, A. (2023). Integrated multi-omics analysis reveals the molecular interplay between circadian clocks and cancer pathogenesis. Scientific Reports, 13(1), 14198. https://doi.org/10.1038/s41598-023-39401-1

Ponce-de-Leon, M., Montagud, A., Noël, V., Meert, A., Pradas, G., Barillot, E., Calzone, L., & Valencia, A. (2023). PhysiBoSS 2.0: A sustainable integration of stochastic Boolean and agent-based modelling frameworks. Npj Systems Biology and Applications, 9(1), 1–12. https://doi.org/10.1038/s41540-023-00314-4

Puccia, C. J., & Levins, R. (1985). Qualitative modeling of complex systems: An introduction to loop analysis and time averaging. Harvard University Press.

Railsback, S. F., & Grimm, V. (2019). Agent-based and individual-based modeling: A practical introduction (2nd ed.). Princeton University Press.

Schelling, T. C. (2006). Micromotives and macrobehavior (OCLC: 916020969). W.W. Norton & Company.

Smaldino, P. E. (2023). Modeling social behavior: Mathematical and agent-based models of social dynamics and cultural evolution. Princeton University Press.

Wilensky, U. (1998). NetLogo flocking model [Computer software]. Center for Connected Learning; Computer Based Modeling at Northwestern University. http://ccl.northwestern.edu/netlogo/models/Flocking

Wilson, R. (2004, October 2). Emergence isn’t magic. New Scientist, 2467. https://www.newscientist.com/letter/mg18424674-700-emergence-isnt-magic/

Thank You!

Linktree

Presentation