{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Monty Hall Problem" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Problem Description:\n", "The Monty Hall Problem is a very famous problem in Probability Theory. The question goes like:\n", "\n", "\n", "Suppose you're on a game show, and you're given the choice of three doors: Behind one door is a car; behind the others, goats. You pick a door, say No. 1, and the host, who knows what's behind the doors, opens another door, say No. 3, which has a goat. He then says to you, \"Do you want to pick door No. 2?\" Is it to your advantage to switch your choice?\n", "\n", "By intution it seems that there shouldn't be any benefit of switching the door. But using Bayes' Theorem we can show that by switching the door the contestant has more chances of winning.\n", "\n", "You can also checkout the wikipedia page: https://en.wikipedia.org/wiki/Monty_Hall_problem" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Probabilistic Interpretetion:\n", "So have 3 random variables Contestant $C \\in \\{1, 2, 3\\}$, Host $H \\in \\{1, 2, 3\\}$ and prize $P \\in \\{1, 2, 3 \\}$. The prize has been put randomly behind the doors therefore: $P(P=1) = P(P=2) = P(P=3) = \\frac{1}{3}$. Also, the contestant is going to choose the door randomly, therefore: $P(C=1) = P(C=2) = P(C=3) = \\frac{1}{3}$. For this problem we can build a Bayesian Network structure like:\n", "\n" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from IPython.display import Image\n", "\n", "Image(\"images/monty.png\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "with the following CPDs:\n", "\n", "
\n",
"\n",
"P(C):\n",
"+----------+----------+-----------+-----------+\n",
"|    C     |     0    |     1     |      2    |\n",
"+----------+----------+-----------+-----------+\n",
"|          |    0.33  |    0.33   |    0.33   |\n",
"+----------+----------+-----------+-----------+\n",
"\n",
"P(P):\n",
"+----------+----------+-----------+-----------+\n",
"|    P     |     0    |     1     |      2    |\n",
"+----------+----------+-----------+-----------+\n",
"|          |    0.33  |    0.33   |    0.33   |\n",
"+----------+----------+-----------+-----------+\n",
"\n",
"P(H | P, C):\n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"|   C  |          0         |          1         |          2         |\n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"|   P  |   0  |   1  |   2  |   0  |   1  |   2  |   0  |   1  |   2  |\n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"|  H=0 |   0  |   0  |   0  |   0  |  0.5 |   1  |   0  |   1  |  0.5 | \n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"|  H=1 |  0.5 |   0  |   1  |   0  |   0  |   0  |   1  |   0  |  0.5 |\n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"|  H=2 |  0.5 |   1  |   0  |   1  |  0.5 |   0  |   0  |   0  |   0  |\n",
"+------+------+------+------+------+------+------+------+------+------+\n",
"
\n", "\n", "Let's say that the contestant selected door 0 and the host opened door 2, we need to find the probability of the prize i.e. $P(P|H=2, C=0)$." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Finding Elimination Order: : : 0it [00:19, ?it/s]\n" ] }, { "data": { "text/plain": [ "[,\n", " ,\n", " ]" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from pgmpy.models import BayesianNetwork\n", "from pgmpy.factors.discrete import TabularCPD\n", "\n", "# Defining the network structure\n", "model = BayesianNetwork([(\"C\", \"H\"), (\"P\", \"H\")])\n", "\n", "# Defining the CPDs:\n", "cpd_c = TabularCPD(\"C\", 3, [[0.33], [0.33], [0.33]])\n", "cpd_p = TabularCPD(\"P\", 3, [[0.33], [0.33], [0.33]])\n", "cpd_h = TabularCPD(\n", " \"H\",\n", " 3,\n", " [\n", " [0, 0, 0, 0, 0.5, 1, 0, 1, 0.5],\n", " [0.5, 0, 1, 0, 0, 0, 1, 0, 0.5],\n", " [0.5, 1, 0, 1, 0.5, 0, 0, 0, 0],\n", " ],\n", " evidence=[\"C\", \"P\"],\n", " evidence_card=[3, 3],\n", ")\n", "\n", "# Associating the CPDs with the network structure.\n", "model.add_cpds(cpd_c, cpd_p, cpd_h)\n", "\n", "# Some other methods\n", "model.get_cpds()" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "True" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# check_model check for the model structure and the associated CPD and returns True if everything is correct otherwise throws an exception\n", "model.check_model()" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "Finding Elimination Order: : : 0it [00:00, ?it/s]\n", "0it [00:00, ?it/s]\u001b[A" ] }, { "name": "stdout", "output_type": "stream", "text": [ "+------+----------+\n", "| P | phi(P) |\n", "+======+==========+\n", "| P(0) | 0.3333 |\n", "+------+----------+\n", "| P(1) | 0.6667 |\n", "+------+----------+\n", "| P(2) | 0.0000 |\n", "+------+----------+\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "\n" ] } ], "source": [ "# Infering the posterior probability\n", "from pgmpy.inference import VariableElimination\n", "\n", "infer = VariableElimination(model)\n", "posterior_p = infer.query([\"P\"], evidence={\"C\": 0, \"H\": 2})\n", "print(posterior_p)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We see that the posterior probability of having the prize behind door 1 is more that door 0. Therefore the contestant should switch the door." ] } ], "metadata": { "anaconda-cloud": {}, "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.10" } }, "nbformat": 4, "nbformat_minor": 4 }