Neural Network Predictions of a Simulation-based Statistical and Graph Theoretic Study of the Board Game RISK

Author

Jacob MunsonFollow

Abstract

We translate the RISK board into a graph which undergoes updates as the game advances. The dissection of the game into a network model in discrete time is a novel approach to examining RISK. A review of the existing statistical findings of skirmishes in RISK is provided. The graphical changes are accompanied by an examination of the statistical properties of RISK. The game is modeled as a discrete time dynamic network graph, with the various features of the game modeled as properties of the network at a given time. As the network is computationally intensive to implement, results are produced by way of computer simulation. We propose three heuristic player strategies of increasing complexity, and demonstrate the effectiveness of each through a series of comparative simulations over a range of scalable values. The features are used to produce a prediction-oriented model based on these findings. The probability of a player win is modeled as a binary response to a single layer feed-forward neural network. We demonstrate the predictive power of our model as well as the performance increase of the player strategies. Recommendations for playing RISK well, based on these results, are given

Year manuscript completed

2017

Year degree awarded

2017

Author's Keywords

Neural Networks, RISK, Statistics, Graph Theory, Simulation, Board Game

Thesis Advisor

Christopher J. Mecklin

Committee Member

Elizabeth A. Donovan

Committee Member

Robert G. Donnelly

Document Type

Thesis

Recommended Citation

Munson, Jacob, "Neural Network Predictions of a Simulation-based Statistical and Graph Theoretic Study of the Board Game RISK" (2017). Murray State Theses and Dissertations. 63.
https://digitalcommons.murraystate.edu/etd/63