Course 4
COURSE 4 · SPECIALIST · 8 TOPICS · TUTORIALS
Simulation, AI & Data Practice
Simulating strategic systems, learning agents, working with live data, and publishing the analysis as reproducible, documented R code.
What you’ll be able to do by the end
- Build Monte Carlo and agent-based simulations and run tournaments between strategies.
- Explain how AlphaZero, counterfactual regret minimisation and multi-agent reinforcement learning relate to game theory.
- Trace machine learning from the perceptron to transformers, with R implementations.
- Retrieve data from public sources such as the World Bank, ACLED, NOAA and arXiv for game-theoretic analysis.
- Produce publication-ready figures, interactive plots, game trees and network graphs.
- Make an analysis reproducible with Quarto, renv, targets and Docker, and package it with roxygen2, testthat and pkgdown.
Who should take this course
For readers who already write R and want to simulate, learn from data, and publish their work reproducibly.
The eight topics
Topic 1 · tutorials
Simulations
Monte Carlo, agent-based models, stochastic simulations, RL agents, tournaments.
Topic 2 · tutorials
ML and Game Theory
AlphaZero, CFR, multi-agent RL, GANs, adversarial robustness.
Topic 3 · tutorials
AI/ML Foundations
From perceptrons to transformers — history, theory, and R implementations.
Topic 4 · tutorials
Public APIs and Datasets
World Bank, ACLED, NOAA, arXiv, Lichess — live data for game-theoretic analysis.
Topic 5 · tutorials
Real-World Data Applications
Empirical game theory with public historical data — trade, conflict, climate, health.
Topic 6 · tutorials
Visualization & Communication
Publication-ready ggplot2, plotly, game trees, network graphs, dashboards.
Topic 7 · tutorials
Reproducibility & Open Science
Quarto, renv, targets, Docker, pre-registration, data sharing.
Topic 8 · tutorials
R Package Development
Package structure, roxygen2, testthat, pkgdown, CRAN submission.
Contents
The other courses
- Course 1 — Foundations of Game Theory — Equilibrium concepts and the canonical games, decisions under uncertainty, evolutionary, cooperative, behavioural and network games, and the history behind them.
- Course 2 — Mechanisms, Behaviour & Ethics — Designing the rules of the game, auctions, behaviour in the lab and the field, incentives in blockchains, ethics, and cases where theory meets reality.
- Course 3 — Mathematical & Statistical Methods — Linear algebra, optimisation, information theory, statistics, Bayesian and causal inference, econometrics and network science: the toolkit behind the models.