Agent Based Modeling and Simulation (ABMS)

  • Dr. Alejandro Guerra Hernández

2025 Edition

Welcome to the course of Agent Based Modeling and Simulation (ABMS) . ABMs can find new, better solutions to many problems important to our environment, heath, and economy. The common feature in these problems is that they occur in systems composed of autonomous agents that interact with each other and their environment. Such agents differ from each other over space and time, and have behaviors that are often very important to how the system works.

The course explores the interaction of ABMS and Machine Learning (ML) techniques. For this, it is organized in five parts, the first four covers ABMS including the basics of NetLogo and it’s use in developing well documented models; the review of different model design concepts; the principles of pattern oriented modeling; and the analysis of the resulting models. The fifth part is about Reinforcement Leaning and its use in ABMs. The organization of the course is as follows:

Introhttps://www.uv.mx/personal/aguerra/wp-admin/options-general.phpduction | abms-syllabus

Agent-Based Modeling and NetLogo Basics

  1. Agent Based Modeling and Simulation | 1:1 | abms-slides-01
  2. Getting Started with NetLogo | 1:2 | abms-slides-02
  3. The ODD Protocol | 1:3 | abms-slides-03
  4. Implementing a First Agent Based Model | 1:4 | abms-slides-04
  5. From Animation to Science | 1:5 | abms-slides-05
  6. Testing your Programs | 1:6 | abms-slides-06

Some Model Design Concepts

  1. Emergence | 1:8 | abms-slides-07
  2. Stochasticity | 1:15 | abms-slides-08
  3. Collectives | 1:16 |abms-slides-09

Pattern-Oriented Modeling

  1. Patterns for Model Structure | 1:17-18 | abms-slides-10
  2. Patterns for Theory Development | 1:19 | abms-slides-11
  3. Patterns for Parameterization and Calibration | 1:20 | abms-slides-12

Model Analysis

  1. Analyzing and Understanding ABMs | 1:22 | abms-slides-13

Reinforcement Learning

  1. Introduction to Reinforcement Learning | 2:1 | abms-slides-14
  2. Finite Markov Decision Processes  | 2:3 | abms-slides-15
  3. Temporal-Difference Learning | 2:6 | abms-slides-16

X:Y stands for reference X, chapter Y.

Ligas

  1. Guía de estilo en NetLogo | URL

References

  1. S. F. Railsback and V. Grimm. Agent-Based and Individual-Based Modeling. Princeton University Press, Princeton, New Jersey, USA, 2019.
  2. R. Sutton and A. G. Barto. Reinforcement Learning: An Introduction. MIT Press, Cambridge, MA., USA, 2018.
Publicado en Páginas