Regret Minimization of Extensive Games and Its Application on Game Strategies

Abstract

Game theory has been discussed by people from generation to generation. The concept of game theory has been applied to various areas. Extensive games, as a typical form of games, are able to simulate the circumstances in many other areas. Recently, more and more attention has been paid to finding a Nash equilibrium in large extensive games. In this paper, we describe a method to solve extensive games, the basis of this method is regret minimization. We then use a concrete case to demonstrate how this technique can be used to make decisions by players in the game of Hearthstone. We calculate the overall regret of all the strategies that the player can take in a given circumstance, and then choose the strategy with the least overall regret. We demonstrate through the case that by minimizing regret in extensive game models, players are able to optimize their strategies and increase the chances to win.