package com.thealgorithms.others;
import java.util.Arrays;
import java.util.Random;
/**
* MiniMax is an algorithm used int artificial intelligence and game theory for
* minimizing the possible loss for the worst case scenario.
*
* See more (https://en.wikipedia.org/wiki/Minimax,
* https://www.geeksforgeeks.org/minimax-algorithm-in-game-theory-set-1-introduction/).
*
* @author aitofi (https://github.com/aitorfi)
*/
public class MiniMaxAlgorithm {
/**
* Game tree represented as an int array containing scores. Each array
* element is a leaf node.
*/
private int[] scores;
private int height;
/**
* Initializes the scores with 8 random leaf nodes
*/
public MiniMaxAlgorithm() {
scores = getRandomScores(3, 99);
height = log2(scores.length);
}
public static void main(String[] args) {
MiniMaxAlgorithm miniMaxAlgorith = new MiniMaxAlgorithm();
boolean isMaximizer = true; // Specifies the player that goes first.
boolean verbose = true; // True to show each players choices.
int bestScore;
bestScore = miniMaxAlgorith.miniMax(0, isMaximizer, 0, verbose);
if (verbose) {
System.out.println();
}
System.out.println(Arrays.toString(miniMaxAlgorith.getScores()));
System.out.println(
"The best score for " + (isMaximizer ? "Maximizer" : "Minimizer") + " is " + String.valueOf(bestScore));
}
/**
* Returns the optimal score assuming that both players play their best.
*
* @param depth Indicates how deep we are into the game tree.
* @param isMaximizer True if it is maximizers turn; otherwise false.
* @param index Index of the leaf node that is being evaluated.
* @param verbose True to show each players choices.
* @return The optimal score for the player that made the first move.
*/
public int miniMax(int depth, boolean isMaximizer, int index, boolean verbose) {
int bestScore, score1, score2;
if (depth == height) { // Leaf node reached.
return scores[index];
}
score1 = miniMax(depth + 1, !isMaximizer, index * 2, verbose);
score2 = miniMax(depth + 1, !isMaximizer, (index * 2) + 1, verbose);
if (isMaximizer) {
// Maximizer player wants to get the maximum possible score.
bestScore = Math.max(score1, score2);
} else {
// Minimizer player wants to get the minimum possible score.
bestScore = Math.min(score1, score2);
}
// Leaf nodes can be sequentially inspected by
// recurssively multiplying (0 * 2) and ((0 * 2) + 1):
// (0 x 2) = 0; ((0 x 2) + 1) = 1
// (1 x 2) = 2; ((1 x 2) + 1) = 3
// (2 x 2) = 4; ((2 x 2) + 1) = 5 ...
if (verbose) {
System.out.println(String.format("From %02d and %02d, %s chooses %02d", score1, score2,
(isMaximizer ? "Maximizer" : "Minimizer"), bestScore));
}
return bestScore;
}
/**
* Returns an array of random numbers which lenght is a power of 2.
*
* @param size The power of 2 that will determine the lenght of the array.
* @param maxScore The maximum possible score.
* @return An array of random numbers.
*/
public static int[] getRandomScores(int size, int maxScore) {
int[] randomScores = new int[(int) Math.pow(2, size)];
Random rand = new Random();
for (int i = 0; i < randomScores.length; i++) {
randomScores[i] = rand.nextInt(maxScore) + 1;
}
return randomScores;
}
// A utility function to find Log n in base 2
private int log2(int n) {
return (n == 1) ? 0 : log2(n / 2) + 1;
}
public void setScores(int[] scores) {
if (scores.length % 1 == 0) {
this.scores = scores;
height = log2(this.scores.length);
} else {
System.out.println("The number of scores must be a power of 2.");
}
}
public int[] getScores() {
return scores;
}
public int getHeight() {
return height;
}
}