{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Week 12 Value function-based Reinforcement Learning Algorithms" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Q1:\n", "Imagine an MDP with two states ($S_1$ and $S_2$) and in which $S_1$ has two actions ($a_1$ and $a_2$) and $S_2$\n", "has only one action ($a_1$). Suppose the discount factor is $\\gamma$ and suppose you run Q-learning with a\n", "Q-table initialized with all zero values and with learning rate $\\alpha$.\n", "- **a)** On the first transition, you start in state $S_1$, apply action $a_1$, receive an immediate reward of\n", "1 and then land in state $S_2$. What is the resulting value of $Q(S_1, a_1)$? Give your answer as\n", "an algebraic expression that may include one or both of the symbols $\\gamma$ and $\\alpha$.\n", "- **b)** On the second transition, you apply action $a_1$, receive an immediate reward of 0 and then\n", "land in state $S_1$. What is the resulting value of $Q(S_2, a_1)$? Give your answer as\n", "an algebraic expression that may include one or both of the symbols $\\gamma$ and $\\alpha$.\n", "- **c)** ) On the third transition, you apply action $a_2$, receive an immediate reward of 1 and then land\n", "in state $S_2$. What is the resulting value of $Q(S_1, a_2)$? Give your answer as\n", "an algebraic expression that may include one or both of the symbols $\\gamma$ and $\\alpha$." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Q-Learning\n", "\n", "Q-Learning is based on the notion of a Q-function. The Q-function (a.k.a the state-action value function) of a policy $\\pi$, $Q^{\\pi}(s, a)$, measures the expected return or discounted sum of rewards obtained from state $s$ by taking action $a$ first and following policy $\\pi$ thereafter. We define the optimal Q-function $Q^*(s, a)$ as the maximum return that can be obtained starting from observation $s$, taking action $a$ and following the optimal policy thereafter. The optimal Q-function obeys the following *Bellman* optimality equation: \n", "\n", "$\\begin{equation}Q^\\ast(s, a) = \\mathbb{E}[ r + \\gamma \\max_{a'} Q^\\ast(s', a') ]\\end{equation}$\n", "\n", "This means that the maximum return from state $s$ and action $a$ is the sum of the immediate reward $r$ and the return (discounted by $\\gamma$) obtained by following the optimal policy thereafter until the end of the episode (i.e., the maximum reward from the next state $s'$). The expectation is computed both over the distribution of immediate rewards $r$ and possible next states $s'$.\n", "\n", "The basic idea behind Q-Learning is to use the Bellman optimality equation as an iterative update $Q_{i+1}(s, a) \\leftarrow \\mathbb{E}\\left[ r + \\gamma \\max_{a'} Q_{i}(s', a')\\right]$, and it can be shown that this converges to the optimal $Q$-function, i.e. $Q_i \\rightarrow Q^*$ as $i \\rightarrow \\infty$ (see the [DQN paper](https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf)).\n", "\n", "### Summing up the Q-Learning Process\n", "In a way, Reinforcement Learning is the science of making optimal decisions using experiences. Breaking it down, the process of Reinforcement Learning involves these simple steps:\n", "\n", "1. Initialize the Q-table by all zeros.\n", "2. Start exploring actions: For each state, select any one among all possible actions for the current state (S).\n", "3. Travel to the next state (S') as a result of that action (a).\n", "4. For all possible actions from the state (S') select the one with the highest Q-value.\n", "5. Update Q-table values using the equation.\n", "6. Set the next state as the current state.\n", "7. If goal state is reached, then end and repeat the process.\n", "\n", "Observation of the environment > Deciding how to act using some strategy > Acting accordingly > Receiving a reward or penalty > Learning from the experiences and refining our strategy/policy > Iterate until an optimal strategy is found" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Practical Tips for Q-learning\n", "- Q-learning takes some care to stablize. Test on easy, reliable tasks first, make sure your implementation is correct.\n", "- Large replay buffers help improve stability.\n", "- It takes time, be patient – might be no better than random for a while.\n", "- Start with high exploration (epsilon) and gradually reduce.\n", "- Bellman error (regression error) gradients can be big; clip gradients or use Huber loss.\n", "- Double Q-learning helps a lot in practice, simple and no downsides.\n", "- Schedule exploration (high to low) and learning rates (high to low), Adam optimizer can help too.\n", "- Run multiple random seeds, it’s very inconsistent between runs." ] }, { "attachments": { "image.png": { "image/png": "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" } }, "cell_type": "markdown", "metadata": {}, "source": [ "## T1: [n-Chain](https://gym.openai.com/envs/NChain-v0/): \n", "The n-Chain environment is taken from the OpenAI Gym module (official documentation),which is a simple 5 state environment.\n", "\n", "This game presents moves along a linear chain of states, with two actions:\n", "* action 0 = move forward along the chain, but get no reward \n", "* action 1 = move backward to state 0, get small reward of 2\n", "\n", "The diagram below demonstrates this environment:\n", "\n", "![image.png](attachment:image.png)\n", "\n", "When action 1 is taken, i.e. move backwards, there is an immediate reward of 2 given to the agent – and the agent is returned to state 0 (back to the beginning of the chain). However, when a move forward action is taken (action 0), there is no immediate reward until state 4. When the agent moves forward while in state 4, a reward of 10 is received by the agent. The agent stays in state 4 at this point also, so the reward can be repeated. There is also a random chance that the agent’s action is “flipped” by the environment (i.e. an action 0 is flipped to an action 1 and vice versa)." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "ename": "ModuleNotFoundError", "evalue": "No module named 'gym'", "output_type": "error", "traceback": [ "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", "\u001b[0;31mModuleNotFoundError\u001b[0m Traceback (most recent call last)", "\u001b[0;32m\u001b[0m in \u001b[0;36m\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;31m# load other necessary pytho modules\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0;32mimport\u001b[0m \u001b[0mgym\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 3\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mnumpy\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mnp\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mmatplotlib\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpyplot\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 5\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mrandom\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n", "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'gym'" ] } ], "source": [ "# load other necessary pytho modules\n", "import gym\n", "import numpy as np\n", "import matplotlib.pyplot as plt\n", "import random\n", "import sys\n", "import warnings\n", "import time\n", "\n", "# ignore warnings\n", "if not sys.warnoptions:\n", " warnings.simplefilter(\"ignore\")" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "# initialize the nchain environment\n", "env = gym.make('NChain-v0')" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[0, 1, 1, 1, 0, 1, 1, 1, 0, 0]" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "# get 10 randomly sampled actions\n", "[env.action_space.sample() for ii in range(10)]" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "# define the class which consists the Q-learning algorithm\n", "class QAgent(object):\n", " \"\"\"\n", " Implementation of a Q-learning Algorithm\n", " \"\"\"\n", " def __init__(self, env, name, state_size, action_size, learning_parameters, exploration_parameters): \n", " \"\"\"\n", " initialize the Q-learning agent\n", " Args:\n", " name (str): set the name of the Q-Agent\n", " state_size (int): ..\n", " action_size (int): ..\n", " learning_parameters (dict): \n", " exploration_parameters (dict):\n", " \"\"\"\n", " self.name = name\n", " \n", " # init the Q-table\n", " self.qtable = np.zeros((state_size, action_size))\n", " self.result = np.zeros((state_size, action_size))\n", " \n", " # learning parameters\n", " self.learning_rate = learning_parameters['learning_rate']\n", " self.gamma = learning_parameters['gamma']\n", " \n", " # exploration parameters\n", " self.epsilon = exploration_parameters['epsilon'] \n", " self.max_epsilon = exploration_parameters['max_epsilon']\n", " self.min_epsilon = exploration_parameters['min_epsilon']\n", " self.decay_rate = exploration_parameters['decay_rate'] \n", " \n", " self.env = env\n", " \n", " \n", " def q_learning(self, plot = False, max_steps = 10, total_episodes = 1000):\n", " \"\"\"\n", " implementation of the q-learning algorithm, here the q-table values are calculated\n", " Args:\n", " plot (boolean): set true, to get trainings progress \n", " max_steps (int): number of stepts an agent can take, before the environment is reset \n", " total_episodes (int): total of training episodes (the number of trials a agent can do) \n", " \"\"\"\n", "\n", " # create placeholders to store the results\n", " self.episode_rewards = np.zeros(total_episodes) \n", " self.episode_epsilon = np.zeros(total_episodes)\n", " self.episode_last_state = np.zeros(total_episodes)\n", " \n", " start = time.time()\n", " # loop over all episodes\n", " for episode_i in range(total_episodes):\n", " # initalize the environment\n", " state = self.env.reset()\n", " \n", " # for each episode loop over the max number of steps that are possible\n", " for step in range(max_steps): \n", " \n", " # get action, e-greedy\n", " action = self.get_action(state)\n", " \n", " # take an action and observe the outcome state (new_state), reward and stopping criterion\n", " new_state, reward, done, _ = self.env.step(action)\n", "\n", " self.qtable[state, action] = self.update_qtable(state, new_state, action, reward, done)\n", " state = new_state\n", " self.episode_rewards[episode_i] += reward \n", " self.result = np.dstack((self.result, self.qtable))\n", "\n", " # check stopping criterion\n", " if done == True:\n", " break\n", " \n", " self.episode_rewards[episode_i] /= step # average the reward\n", " self.episode_last_state[episode_i] = state # average the reward\n", " \n", " # reduce epsilon, for exploration-exploitation tradeoff\n", " self.update_epsilon(episode_i)\n", " \n", " if episode_i % 100 == 0 and plot:\n", " print('episode: {}'.format(episode_i))\n", " print('\\telapsed time [min]: {}'.format(round( (time.time() - start)/60, 1)))\n", "\n", " \n", " def update_qtable(self, state, new_state, action, reward, done):\n", " \"\"\"\n", " update the q-table: Q(s,a) = Q(s,a) + lr * [R(s,a) + gamma * max Q(s',a') - Q (s,a)]\n", " Args:\n", " state (int): current state of the environment\n", " new_state (int): new state of the environment\n", " action (int): current action taken by agent\n", " reward (int): current reward received from env\n", " done (boolean): variable indicating if env is done\n", " Returns:\n", " qtable (array): the qtable containing a value for every state (y-axis) and action (x-axis) \n", " \"\"\"\n", " return self.qtable[state,action] + self.learning_rate * \\\n", " (reward + self.gamma * np.max(self.qtable[new_state, :]) * (1- done) - self.qtable[state, action])\n", " \n", " \n", " def update_epsilon(self, episode):\n", " \"\"\"\n", " reduce epsilon, exponential decay\n", " Args:\n", " episode (int): number of episode\n", " \"\"\"\n", " self.epsilon = self.min_epsilon + (self.max_epsilon - self.min_epsilon) * np.exp(-self.decay_rate*episode)\n", "\n", " \n", " def get_action(self, state):\n", " \"\"\"\n", " select action e-greedy\n", " Args:\n", " state (int): current state of the environment/agent\n", " Returns:\n", " action (int): action that the agent will take in the next step\n", " \"\"\"\n", " if random.uniform(0,1) >= self.epsilon:\n", " # exploitation, max value for given state\n", " action = np.argmax(self.qtable[state, :])\n", " else:\n", " # exploration, random choice\n", " action = self.env.action_space.sample()\n", " return action" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "If we think about the previous iteration of the agent training model using Q learning, the action selection policy is based solely on the maximum Q value in any given state. It is conceivable that, given the random nature of the environment, that the agent initially makes “bad” decisions. The Q values arising from these decisions may easily be “locked in” – and from that time forward, bad decisions may continue to be made by the agent because it can only ever select the maximum Q value in any given state, even if these values are not necessarily optimal. This action selection policy is called a greedy policy.\n", "\n", "So we need a way for the agent to eventually always choose the “best” set of actions in the environment, yet at the same time allowing the agent to not get “locked in” and giving it some space to explore alternatives. What is required is the $\\epsilon$-greedy policy.\n", "\n", "The $\\epsilon$-greedy policy in reinforcement learning is basically the same as the greedy policy, except that there is a value $\\epsilon$ (which may be set to decay over time) where, if a random number is selected which is less than this value, an action is chosen completely at random. This step allows some random exploration of the value of various actions in various states, and can be scaled back over time to allow the algorithm to concentrate more on exploiting the best strategies that it has found" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [], "source": [ "# avoid slipping in on the chain\n", "env.env.slip = 0\n", "\n", "action_size = env.action_space.n\n", "state_size = env.observation_space.n\n", "\n", "# Set the Training parameters\n", "# set true, to get printed out the trainings progress\n", "plot = False \n", "#plot = True\n", "# Set number of stepts an agent can take, before the environment is reset, \n", "max_steps = 10 \n", "# Set total of training episodes (the number of trials a agent can do) \n", "total_episodes = 1000" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "After enough random exploration of actions, the Q-values tend to converge serving our agent as an action-value function which it can exploit to pick the most optimal action from a given state.\n", "\n", "There's a tradeoff between exploration (choosing a random action) and exploitation (choosing actions based on already learned Q-values). We want to prevent the action from always taking the same route, and possibly overfitting, so we'll be introducing another parameter called ϵ \"epsilon\" to cater to this during training.\n", "\n", "Instead of just selecting the best learned Q-value action, we'll sometimes favor exploring the action space further. Lower epsilon value results in episodes with more penalties (on average) which is obvious because we are exploring and making random decisions." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Agent 1\n", "Create an agent that explores and takes future rewards into account" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "### case1\n", "name = 'case1, the agent explores and takes future rewards into account'\n", "\n", "# q-learning parameters\n", "learning_parameters = {\n", " 'learning_rate': 0.8,\n", " 'gamma': 0.9} # gamma = 1, we care about all future rewards equally as the current one \n", " # gamma = 0, we only care about the current reward)\n", "\n", "# exploration-exploitation parameters\n", "exploration_parameters = {\n", " 'epsilon': 1,\n", " 'max_epsilon': 1,\n", " 'min_epsilon': 0.0,\n", " 'decay_rate': 0.008} # smaller decay rate, more exploration\n", "\n", "q_agent_1 = QAgent(env, name, state_size, action_size, learning_parameters, exploration_parameters)\n", "q_agent_1.q_learning(plot = plot, max_steps = max_steps, total_episodes = total_episodes)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Agent 2\n", "Create an agent that cares only about immediate rewards (gamma is very small)" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "### case2\n", "name = 'case2, the agent cares only about immediate rewards (small gamma)'\n", "\n", "# q-learning parameters\n", "learning_parameters = {\n", " 'learning_rate': 0.8,\n", " 'gamma': 0.01} # gamma = 1, we care about all future rewards equally as the current one \n", " # gamma = 0, we only care about the current reward)\n", "\n", "# exploration-exploitation parameters\n", "exploration_parameters = {\n", " 'epsilon': 1,\n", " 'max_epsilon': 0.5,\n", " 'min_epsilon': 0.0,\n", " 'decay_rate': 0.008} # smaller decay rate, more exploration\n", "\n", "\n", "q_agent_2 = QAgent(env, name, state_size, action_size, learning_parameters, exploration_parameters)\n", "q_agent_2.q_learning(plot = plot, max_steps = max_steps, total_episodes = total_episodes)\n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Agent 3\n", "Create an agent that doesn't explore the environment (small epsilon)" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "### case3\n", "name = \"case3, the agent doesn't explore the environment (small epsilon)\"\n", "\n", "# q-learning parameters\n", "learning_parameters = {\n", " 'learning_rate': 0.8,\n", " 'gamma': 0.9} # gamma = 1, we care about all future rewards equally as the current one \n", " # gamma = 0, we only care about the current reward)\n", "\n", "# exploration-exploitation parameters\n", "exploration_parameters = {\n", " 'epsilon': 1,\n", " 'max_epsilon': 0.2,\n", " 'min_epsilon': 0.0,\n", " 'decay_rate': 0.5} # smaller decay rate, more exploration\n", "\n", "\n", "q_agent_3 = QAgent(env, name, state_size, action_size, learning_parameters, exploration_parameters)\n", "q_agent_3.q_learning(plot = plot, max_steps = max_steps, total_episodes = total_episodes)" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "def visualize_episodes(*cases):\n", " \"\"\"\n", " create a visualization method\n", " Args:\n", " cases (object): trained q-learning agent\n", " \"\"\"\n", "\n", " plt.figure(figsize=(10,15))\n", " colour = ['orange', 'm','b','k','g']\n", " \n", " # 1\n", " plt.subplot(3,1,1)\n", " plt.title('Reward over time per episode')\n", " for i, case in enumerate(cases):\n", " plt.plot(case.episode_rewards, c = colour[i], label = case.name, \n", " linewidth = 1, linestyle = '-', alpha = 0.7)\n", " plt.xlabel('# episodes')\n", " plt.ylabel('Reward')\n", " plt.grid()\n", " plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05),\n", " fancybox=True, shadow=True, ncol=1)\n", " \n", " # 2\n", " plt.subplot(3,1,2)\n", " plt.title('Exploration parameter epsilon per episode')\n", " for i, case in enumerate(cases):\n", " plt.plot(case.episode_epsilon, c = colour[i], label = case.name, \n", " linewidth = 1, linestyle = '-', alpha = 0.7)\n", " plt.xlabel('# episodes')\n", " plt.ylabel('Epsilon')\n", " plt.grid()\n", " plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05),\n", " fancybox=True, shadow=True, ncol=1) \n", " \n", " # 3\n", " plt.subplot(3,1,3)\n", " plt.title('Last state the agent is standing on at the end of the episode')\n", " for i, case in enumerate(cases):\n", " plt.plot(case.episode_last_state, c = colour[i], label = case.name, \n", " linewidth = 1, linestyle = '-', alpha = 0.7)\n", " plt.xlabel('# episodes')\n", " plt.ylabel('state number')\n", " plt.grid()\n", " plt.legend(loc='upper center', bbox_to_anchor=(0.5, -0.05),\n", " fancybox=True, shadow=True, ncol=1) \n", " plt.tight_layout()\n", " plt.show()\n", " " ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "# visualize the different set of parameters\n", "visualize_episodes(q_agent_1, q_agent_2, q_agent_3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "* y-axis: state 0 to state 4\n", "* x-axis: action 0 and action 1\n", "* action 0 = move forward, but get no reward (in the last state get large reward)\n", "* action 1 = move backward to state 0, get small reward\n", "### higher values mean higher future rewards for this specific action-state pair" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "case1:\n", " [[ 65.6 61. ]\n", " [ 72.9 61. ]\n", " [ 81. 61. ]\n", " [ 90. 61. ]\n", " [100. 61. ]],\n", "\n", "case2:\n", " [[ 0. 2. ]\n", " [ 0. 2. ]\n", " [ 0. 2. ]\n", " [ 0.1 0. ]\n", " [10.1 2. ]],\n", "\n", "case3:\n", " [[ 4.1 20. ]\n", " [ 0. 4.7]\n", " [ 0. 3.8]\n", " [ 0. 0. ]\n", " [ 0. 0. ]]\n" ] } ], "source": [ "print('case1:\\n {},\\n\\ncase2:\\n {},\\n\\ncase3:\\n {}'.format(np.around(q_agent_1.qtable,1),\n", " np.around(q_agent_2.qtable,1),\n", " np.around(q_agent_3.qtable,1))) \n" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The q-table of case1 should have similar values like the following table:\n", "\n", "| ---- | Action 0 | Action 1 | \n", "| ------------- | ------------- | ------------- |\n", "| state 0 | 65.61 | 61.049 |\n", "| state 1 | 72.9 | 61.049 |\n", "| state 2 | 81. | 61.049 |\n", "| state 3 | 90. | 61.049 |\n", "| state 4 | 100. | 61.049 |" ] }, { "attachments": { "image.png": { "image/png": "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" } }, "cell_type": "markdown", "metadata": {}, "source": [ "## T2: Self-Driving Cab via RL\n", "Let's design a simulation of a self-driving cab. The major goal is to demonstrate, in a simplified environment, how you can use RL techniques to develop an efficient and safe approach for tackling this problem.\n", "\n", "The Smartcab's job is to pick up the passenger at one location and drop them off in another. Here are a few things that we'd love our Smartcab to take care of:\n", "1. Drop off the passenger to the right location.\n", "2. Save passenger's time by taking minimum time possible to drop off\n", "3. Take care of passenger's safety and traffic rules\n", "4. There are different aspects that need to be considered here while modeling an RL solution to this problem: rewards, states, and actions.\n", "\n", "### Rewards\n", "1. The agent should receive a high positive reward for a successful dropoff because this behavior is highly desired\n", "2. The agent should be penalized if it tries to drop off a passenger in wrong locations\n", "3. The agent should get a slight negative reward for not making it to the destination after every time-step. \"Slight\" negative because we would prefer our agent to reach late instead of making wrong moves trying to reach to the destination as fast as possible\n", "\n", "### State Space\n", "In Reinforcement Learning, the agent encounters a state, and then takes action according to the state it's in.\n", "\n", "The State Space is the set of all possible situations our taxi could inhabit. The state should contain useful information the agent needs to make the right action.\n", "\n", "Let's say we have a training area for our Smartcab where we are teaching it to transport people in a parking lot to four different locations (R, G, Y, B):\n", "![image.png](attachment:image.png)\n", "Let's assume Smartcab is the only vehicle in this parking lot. We can break up the parking lot into a 5x5 grid, which gives us 25 possible taxi locations. These 25 locations are one part of our state space. Notice the current location state of our taxi is coordinate (3, 1).\n", "\n", "You'll also notice there are four (4) locations that we can pick up and drop off a passenger: R, G, Y, B or [(0,0), (0,4), (4,0), (4,3)] in (row, col) coordinates. Our illustrated passenger is in location Y and they wish to go to location R.\n", "\n", "When we also account for one (1) additional passenger state of being inside the taxi, we can take all combinations of passenger locations and destination locations to come to a total number of states for our taxi environment; there's four (4) destinations and five (4 + 1) passenger locations.\n", "\n", "So, our taxi environment has 5×5×5×4=500 total possible states.\n", "\n", "### Action Space\n", "The agent encounters one of the 500 states and it takes an action. The action in our case can be to move in a direction or decide to pickup/dropoff a passenger.\n", "\n", "In other words, we have six possible actions:\n", "\n", "1. south\n", "2. north\n", "3. east\n", "4. west\n", "5. pickup\n", "6. dropoff\n", "This is the action space: the set of all the actions that our agent can take in a given state.\n", "\n", "You'll notice in the illustration above, that the taxi cannot perform certain actions in certain states due to walls. In environment's code, we will simply provide penalty for every wall hit and the taxi won't move anywhere. This will just rack up penalties causing the taxi to consider going around the wall.\n", "\n", "### Q-Table\n", "The Q-table is a matrix where we have a row for every state (500) and a column for every action (6). It's first initialized to 0, and then values are updated after training. Note that the Q-table has the same dimensions as the reward table, but it has a completely different purpose.\n", "\n", "Q-Table values are initialized to zero and then updated during training to values that optimize the agent's traversal through the environment for maximum rewards" ] }, { "cell_type": "code", "execution_count": 13, "metadata": {}, "outputs": [], "source": [ "#load the necessary pytho modules\n", "import numpy as np\n", "import gym\n", "import random" ] }, { "cell_type": "code", "execution_count": 32, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "+---------+\n", "|R: | : :G|\n", "| : | : : |\n", "| : : : : |\n", "| | : | : |\n", "|\u001b[34;1m\u001b[43mY\u001b[0m\u001b[0m| : |\u001b[35mB\u001b[0m: |\n", "+---------+\n", "\n" ] } ], "source": [ "#load the game environment and render what it looks like:\n", "env = gym.make(\"Taxi-v3\")\n", "env.render()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "1. The filled square represents the taxi, which is yellow without a passenger and green with a passenger.\n", "2. The pipe (\"|\") represents a wall which the taxi cannot cross.\n", "3. R, G, Y, B are the possible pickup and destination locations. The blue letter represents the current passenger pick-up location, and the purple letter is the current destination." ] }, { "cell_type": "code", "execution_count": 33, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "+---------+\n", "|R: | : :\u001b[35mG\u001b[0m|\n", "| : | : : |\n", "| : : : : |\n", "| | : | : |\n", "|\u001b[34;1mY\u001b[0m|\u001b[43m \u001b[0m: |B: |\n", "+---------+\n", "\n", "Action Space Discrete(6)\n", "State Space Discrete(500)\n" ] } ], "source": [ "env.reset() # reset environment to a new, random state\n", "env.render()\n", "\n", "print(\"Action Space {}\".format(env.action_space))\n", "print(\"State Space {}\".format(env.observation_space))" ] }, { "cell_type": "code", "execution_count": 34, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "State: 328\n", "+---------+\n", "|R: | : :\u001b[35mG\u001b[0m|\n", "| : | : : |\n", "| : : : : |\n", "| | : | : |\n", "|\u001b[34;1mY\u001b[0m|\u001b[43m \u001b[0m: |B: |\n", "+---------+\n", "\n" ] } ], "source": [ "state = env.encode(3, 1, 2, 0) # (taxi row, taxi column, passenger index, destination index)\n", "print(\"State:\", state)\n", "\n", "env.s = state\n", "env.render()" ] }, { "cell_type": "code", "execution_count": 35, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{0: [(1.0, 428, -1, False)],\n", " 1: [(1.0, 228, -1, False)],\n", " 2: [(1.0, 348, -1, False)],\n", " 3: [(1.0, 328, -1, False)],\n", " 4: [(1.0, 328, -10, False)],\n", " 5: [(1.0, 328, -10, False)]}" ] }, "execution_count": 35, "metadata": {}, "output_type": "execute_result" } ], "source": [ "env.P[328]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This dictionary has the structure {action: [(probability, nextstate, reward, done)]}" ] }, { "cell_type": "code", "execution_count": 36, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Timesteps taken: 200\n", "Penalties incurred: 62\n" ] } ], "source": [ "env.s = 328 # set environment to illustration's state\n", "\n", "epochs = 0\n", "penalties, reward = 0, 0\n", "\n", "frames = [] # for animation\n", "\n", "done = False\n", "\n", "while not done:\n", " action = env.action_space.sample()\n", " state, reward, done, info = env.step(action)\n", "\n", " if reward == -10:\n", " penalties += 1\n", " \n", " # Put each rendered frame into dict for animation\n", " frames.append({\n", " 'frame': env.render(mode='ansi'),\n", " 'state': state,\n", " 'action': action,\n", " 'reward': reward\n", " }\n", " )\n", "\n", " epochs += 1\n", " \n", " \n", "print(\"Timesteps taken: {}\".format(epochs))\n", "print(\"Penalties incurred: {}\".format(penalties))" ] }, { "cell_type": "code", "execution_count": 37, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "+---------+\n", "|R: |\u001b[43m \u001b[0m: :\u001b[35mG\u001b[0m|\n", "| : | : : |\n", "| : : : : |\n", "| | : | : |\n", "|\u001b[34;1mY\u001b[0m| : |B: |\n", "+---------+\n", " (West)\n", "\n", "Timestep: 200\n", "State: 49\n", "Action: 3\n", "Reward: -1\n" ] } ], "source": [ "from IPython.display import clear_output\n", "from time import sleep\n", "\n", "def print_frames(frames):\n", " for i, frame in enumerate(frames):\n", " clear_output(wait=True)\n", " print(frame['frame'])\n", " print(f\"Timestep: {i + 1}\")\n", " print(f\"State: {frame['state']}\")\n", " print(f\"Action: {frame['action']}\")\n", " print(f\"Reward: {frame['reward']}\") \n", " sleep(.1)\n", " \n", "print_frames(frames)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Then, we'll initialize the Q-table to a 500×6 matrix of zeros:" ] }, { "cell_type": "code", "execution_count": 38, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[[0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0.]\n", " ...\n", " [0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0.]\n", " [0. 0. 0. 0. 0. 0.]]\n" ] } ], "source": [ "q_table = np.zeros([env.observation_space.n, env.action_space.n])\n", "print(q_table)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We can now create the training algorithm that will update this Q-table as the agent explores the environment over thousands of episodes." ] }, { "cell_type": "code", "execution_count": 39, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Episode: 100000\n", "Training finished.\n", "\n", "Wall time: 44.7 s\n" ] } ], "source": [ "%%time\n", "\"\"\"Training the agent\"\"\"\n", "\n", "#import random\n", "from IPython.display import clear_output\n", "\n", "# Hyperparameters\n", "alpha = 0.1\n", "gamma = 0.6\n", "epsilon = 0.1\n", "\n", "# For plotting metrics\n", "all_epochs = []\n", "all_penalties = []\n", "\n", "for i in range(1, 100001):\n", " state = env.reset()\n", "\n", " epochs, penalties, reward, = 0, 0, 0\n", " done = False\n", " \n", " while not done:\n", " if random.uniform(0, 1) < epsilon:\n", " action = env.action_space.sample() # Explore action space\n", " else:\n", " action = np.argmax(q_table[state]) # Exploit learned values\n", "\n", " next_state, reward, done, info = env.step(action) \n", " \n", " old_value = q_table[state, action]\n", " next_max = np.max(q_table[next_state])\n", " \n", " new_value = (1 - alpha) * old_value + alpha * (reward + gamma * next_max)\n", " q_table[state, action] = new_value\n", "\n", " if reward == -10:\n", " penalties += 1\n", "\n", " state = next_state\n", " epochs += 1\n", " \n", " if i % 100 == 0:\n", " clear_output(wait=True)\n", " print(f\"Episode: {i}\")\n", "\n", "print(\"Training finished.\\n\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's evaluate the performance of our agent. We don't need to explore actions any further, so now the next action is always selected using the best Q-value:" ] }, { "cell_type": "code", "execution_count": 40, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Results after 100 episodes:\n", "Average timesteps per episode: 12.78\n", "Average penalties per episode: 0.0\n" ] } ], "source": [ "\"\"\"Evaluate agent's performance after Q-learning\"\"\"\n", "\n", "total_epochs, total_penalties = 0, 0\n", "episodes = 100\n", "frames = [] # for animation\n", "for _ in range(episodes):\n", " state = env.reset()\n", " epochs, penalties, reward = 0, 0, 0\n", " \n", " done = False\n", " \n", " while not done:\n", " action = np.argmax(q_table[state])\n", " state, reward, done, info = env.step(action)\n", "\n", " if reward == -10:\n", " penalties += 1\n", "\n", " # Put each rendered frame into dict for animation\n", " frames.append({\n", " 'frame': env.render(mode='ansi'),\n", " 'state': state,\n", " 'action': action,\n", " 'reward': reward\n", " }\n", " )\n", " epochs += 1\n", "\n", " total_penalties += penalties\n", " total_epochs += epochs\n", "\n", "print(f\"Results after {episodes} episodes:\")\n", "print(f\"Average timesteps per episode: {total_epochs / episodes}\")\n", "print(f\"Average penalties per episode: {total_penalties / episodes}\")" ] }, { "cell_type": "code", "execution_count": 41, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "+---------+\n", "|R: | : :G|\n", "| : | : : |\n", "| : : : : |\n", "| | : | : |\n", "|\u001b[35m\u001b[34;1m\u001b[43mY\u001b[0m\u001b[0m\u001b[0m| : |B: |\n", "+---------+\n", " (Dropoff)\n", "\n", "Timestep: 1278\n", "State: 410\n", "Action: 5\n", "Reward: 20\n" ] } ], "source": [ "print_frames(frames)" ] } ], "metadata": { "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": "3.6.4" } }, "nbformat": 4, "nbformat_minor": 2 }