OpenAI Gym<\/a>\uc740 \uac15\ud654\ud559\uc2b5 \uc5d0\uc774\uc804\ud2b8\ub97c \uac1c\ubc1c\ud558\uace0 \ud3c9\uac00\ud560 \uc218 \uc788\ub294 \ud658\uacbd\uc744 \uc81c\uacf5\ud558\ub294 \ud234\ud0b7\uc785\ub2c8\ub2e4. \ub2e4\uc591\ud55c \ud658\uacbd\uc744 \uad6c\uc131\ud558\uace0 \uc0d8\ud50c\ub9c1\ud560 \uc218 \uc788\ub3c4\ub85d \ud574\uc8fc\uc5b4, \uc5f0\uad6c\uc790 \ubc0f \uac1c\ubc1c\uc790\ub4e4\uc774 \uc27d\uac8c \uc2e4\ud5d8\ud574\ubcfc \uc218 \uc788\ub294 \ud1a0\ub300\ub97c \ub9c8\ub828\ud569\ub2c8\ub2e4.<\/p>\nOpenAI Gym\uc5d0\ub294 \ub2e4\uc591\ud55c \ud658\uacbd\uc774 \ud3ec\ud568\ub418\uc5b4 \uc788\uc73c\uba70, ‘Atari \uac8c\uc784’, ‘\ub85c\ubd07 \uc2dc\ubbac\ub808\uc774\uc158’, ‘\ubb38\uc81c \ud574\uacb0’ \ub4f1 \ub2e4\uc591\ud55c \uc885\ub958\uc758 \ubb38\uc81c\uac00 \uc815\uc758\ub418\uc5b4 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \uc5d0\uc774\uc804\ud2b8\uac00 \ub2e4\uc591\ud55c \uc0c1\ud669\uc5d0\uc11c \uc5b4\ub5bb\uac8c \ud559\uc2b5\ud558\ub294\uc9c0 \ud14c\uc2a4\ud2b8\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<\/section>\n\n3. Python \ud658\uacbd \uc124\uc815<\/h2>\n
Python\uc744 \uc0ac\uc6a9\ud558\uc5ec \uac15\ud654\ud559\uc2b5 \ud658\uacbd\uc744 \uad6c\ucd95\ud558\uae30 \uc704\ud574 \uae30\ubcf8\uc801\uc73c\ub85c \ud544\uc694\ud55c \ud328\ud0a4\uc9c0\ub97c \uc124\uce58\ud574\uc57c \ud569\ub2c8\ub2e4. \uc5ec\uae30\uc5d0\uc11c\ub294 OpenAI Gym<\/strong>, NumPy<\/strong>, Matplotlib<\/strong> \ub4f1\uc744 \uc0ac\uc6a9\ud560 \uac83\uc785\ub2c8\ub2e4.<\/p>\n\uc544\ub798\ub294 \ud658\uacbd\uc744 \uc124\uc815\ud558\uae30 \uc704\ud55c \uae30\ubcf8\uc801\uc778 \uc124\uce58 \uba85\ub839\uc5b4\uc785\ub2c8\ub2e4:<\/p>\n
pip install gym numpy matplotlib<\/code><\/pre>\n<\/section>\n\n4. OpenAI Gym\uc744 \ud65c\uc6a9\ud55c \ud658\uacbd \uad6c\ucd95<\/h2>\n
OpenAI Gym\uc744 \uc0ac\uc6a9\ud558\uc5ec \ub2e8\uc21c\ud55c \ud658\uacbd\uc744 \uad6c\ucd95\ud558\ub294 \uc608\uc81c\ub97c \uc0b4\ud3b4\ubcf4\uaca0\uc2b5\ub2c8\ub2e4. \uc774 \uc608\uc81c\uc5d0\uc11c\ub294 ‘CartPole’ \ud658\uacbd\uc744 \uc0ac\uc6a9\ud569\ub2c8\ub2e4. ‘CartPole’\uc740 \ub9c9\ub300\uac00 \uc218\uc9c1\uc73c\ub85c \uc11c \uc788\uc73c\uba74 \ubc8c\uc5b4\uc9c0\ub294 \uc77c\ub828\uc758 \uc0c1\ud638\uc791\uc6a9\uc744 \ud3ec\ud568\ud569\ub2c8\ub2e4. \ubaa9\ud45c\ub294 \uc774\ub97c \uc9c0\uc18d\uc801\uc73c\ub85c \uc720\uc9c0\ud558\ub294 \uac83\uc785\ub2c8\ub2e4.<\/p>\n
\nimport gym\n\n# CartPole \ud658\uacbd \uc0dd\uc131\nenv = gym.make('CartPole-v1')\n\n# \ud658\uacbd \ucd08\uae30\ud654\nstate = env.reset()\n\ndone = False\nwhile not done:\n # \ub79c\ub364\ud55c \ud589\ub3d9 \uc120\ud0dd\n action = env.action_space.sample() \n # \ud658\uacbd\uc5d0\uc11c \ud589\ub3d9\uc744 \ucde8\ud558\uace0 \uacb0\uacfc\ub97c \uad00\ucc30\n next_state, reward, done, info = env.step(action)\n \n # \ud604\uc7ac \uc0c1\ud0dc\ub97c \ub2e4\uc74c \uc0c1\ud0dc\ub85c \uc5c5\ub370\uc774\ud2b8\n state = next_state\n\n# \ud658\uacbd \uc885\ub8cc\nenv.close()\n <\/code><\/pre>\n\uc704 \ucf54\ub4dc\ub294 \ubb34\uc791\uc704 \ud589\ub3d9\uc744 \uc120\ud0dd\ud558\uc5ec ‘CartPole’ \ud658\uacbd\uc744 \ud0d0\uc0c9\ud558\ub294 \uac04\ub2e8\ud55c \uc608\uc81c\uc785\ub2c8\ub2e4. \uc774 \ucf54\ub4dc\ub97c \uc2e4\ud589\ud558\uba74 \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uacfc \uc0c1\ud638\uc791\uc6a9\ud558\ub294 \uacfc\uc815\uc744 \ubcfc \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<\/section>\n\n5. \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5<\/h2>\n
\uc5d0\uc774\uc804\ud2b8\ub97c \ud559\uc2b5\uc2dc\ud0a4\uae30 \uc704\ud574\uc11c\ub294 RL \uc54c\uace0\ub9ac\uc998\uc744 \uc0ac\uc6a9\ud574\uc57c \ud569\ub2c8\ub2e4. \uc5ec\uae30\uc11c\ub294 \uac00\uc7a5 \ub110\ub9ac \uc0ac\uc6a9\ub418\ub294 Q-\ub7ec\ub2dd(Q-learning) \uc54c\uace0\ub9ac\uc998\uc758 \uae30\ubcf8 \uac1c\ub150\uacfc \uc774\ub97c \uad6c\ud604\ud558\ub294 \ubc29\ubc95\uc744 \uc0b4\ud3b4\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n
Q-\ub7ec\ub2dd\uc740 \ud2b9\uc815 \uc0c1\ud0dc-\ud589\ub3d9 \uc30d\uc758 \uac00\uce58\ub97c \ucd94\uc815\ud558\uc5ec \ucd5c\uc801 \uc815\ucc45\uc744 \ud559\uc2b5\ud558\ub294 \uc624\ud504\ub77c\uc778 \ubca8\ub9cc \ubc29\uc815\uc2dd \uae30\ubc18\uc758 \ud559\uc2b5 \uc54c\uace0\ub9ac\uc998\uc785\ub2c8\ub2e4. Q-\ub7ec\ub2dd\uc758 \ud575\uc2ec\uc740 Q-\uac12(Q-value)\uc744 \uc5c5\ub370\uc774\ud2b8\ud558\ub294 \uac83\uc785\ub2c8\ub2e4. \uc774\ub97c \uc704\ud574 \ub2e4\uc74c\uacfc \uac19\uc740 \uc5c5\ub370\uc774\ud2b8 \uacf5\uc2dd\uc744 \uc0ac\uc6a9\ud569\ub2c8\ub2e4:<\/p>\n
\nQ(s, a) \u2190 Q(s, a) + \u03b1 [R(s, a) + \u03b3 max a' Q(s', a') - Q(s, a)]\n <\/code><\/pre>\n\uc5ec\uae30\uc11c \uac01\uac01\uc758 \ubcc0\uc218\ub294 \ub2e4\uc74c\uacfc \uac19\uc740 \uc758\ubbf8\ub97c \uac00\uc9d1\ub2c8\ub2e4:<\/p>\n
\n- \u03b1: \ud559\uc2b5\ub960 (Learning Rate)<\/li>\n
- R: \ubcf4\uc0c1 \ud568\uc218 (Reward Function)<\/li>\n
- \u03b3: \ud560\uc778\uc728 (Discount Factor)<\/li>\n
- Q(s, a): \ud604\uc7ac \uc0c1\ud0dc s\uc5d0\uc11c \ud589\ub3d9 a\uc758 Q-\uac12<\/li>\n<\/ul>\n
\uc544\ub798\uc758 \ucf54\ub4dc\ub294 Q-\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998\uc744 \uc774\uc6a9\ud558\uc5ec ‘CartPole’ \ud658\uacbd\uc5d0\uc11c \uc5d0\uc774\uc804\ud2b8\ub97c \ud559\uc2b5\uc2dc\ud0a4\ub294 \uc608\uc81c\ub97c \ubcf4\uc5ec\uc90d\ub2c8\ub2e4:<\/p>\n
\nimport numpy as np\nimport gym\n\n# Q-\ud14c\uc774\ube14 \ucd08\uae30\ud654\nenv = gym.make('CartPole-v1')\nq_table = np.zeros((env.observation_space.shape[0], env.action_space.n))\n\n# Hyperparameters\nlearning_rate = 0.1\ndiscount_factor = 0.95\nnum_episodes = 1000\n\nfor episode in range(num_episodes):\n state = env.reset()\n done = False\n\n while not done:\n # \ud589\ub3d9 \uc120\ud0dd (Epsilon-greedy \uc815\ucc45)\n if np.random.uniform(0, 1) < 0.1:\n action = env.action_space.sample() # Exploration\n else:\n action = np.argmax(q_table[state]) # Exploitation\n \n # \ud658\uacbd\uc5d0\uc11c \ud589\ub3d9\uc744 \ucde8\ud558\uace0 \ub2e4\uc74c \uc0c1\ud0dc \ubc0f \ubcf4\uc0c1\uc744 \ubc1b\uc74c\n next_state, reward, done, _ = env.step(action)\n \n # Q-\uac12 \uc5c5\ub370\uc774\ud2b8\n q_table[state, action] = q_table[state, action] + learning_rate * (reward + discount_factor * np.max(q_table[next_state]) - q_table[state, action])\n \n # \uc0c1\ud0dc\ub97c \ub2e4\uc74c \uc0c1\ud0dc\ub85c \uc5c5\ub370\uc774\ud2b8\n state = next_state\n\nenv.close()\n <\/code><\/pre>\n\uc704 \ucf54\ub4dc\ub294 \uac04\ub2e8\ud55c Q-\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ud604\ud558\ub294 \uc608\uc81c\uc785\ub2c8\ub2e4. \uc774 \uc608\uc81c\ub97c \ud1b5\ud574 \uc6b0\ub9ac\ub294 \uc5d0\uc774\uc804\ud2b8\uac00 ‘CartPole’ \ud658\uacbd\uc5d0\uc11c \ucd5c\uc801\uc758 \ud589\ub3d9\uc744 \ud559\uc2b5\ud558\ub294 \uacfc\uc815\uc744 \ubcfc \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n<\/section>\n\n6. \uacb0\uacfc \uc2dc\uac01\ud654 \ubc0f \uc131\ub2a5 \ud3c9\uac00<\/h2>\n
\ud6c8\ub828 \ud6c4 \uc5d0\uc774\uc804\ud2b8\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud558\uace0 \uacb0\uacfc\ub97c \uc2dc\uac01\ud654\ud558\ub294 \uac83\uc740 \uc911\uc694\ud55c \uacfc\uc815\uc785\ub2c8\ub2e4. \uc2dc\uac01\ud654\ub97c \ud1b5\ud574 \uc5d0\uc774\uc804\ud2b8\uc758 \ud559\uc2b5 \uc9c4\ud589 \uc0c1\ud669\uc744 \uc9c1\uad00\uc801\uc73c\ub85c \uc774\ud574\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc544\ub798 \ucf54\ub4dc\ub294 \ud6c8\ub828\ub41c \uc5d0\uc774\uc804\ud2b8\uc758 \uc131\ub2a5\uc744 \uc2dc\uac01\ud654\ud558\ub294 \uc608\uc81c\uc785\ub2c8\ub2e4:<\/p>\n
\nimport matplotlib.pyplot as plt\n\n# \uc5d0\ud53c\uc18c\ub4dc\uc640 \ubcf4\uc0c1 \uc800\uc7a5\nreward_list = []\n\nfor episode in range(100):\n state = env.reset()\n total_reward = 0\n done = False\n \n while not done:\n action = np.argmax(q_table[state])\n state, reward, done, _ = env.step(action)\n total_reward += reward\n \n reward_list.append(total_reward)\n\nplt.plot(reward_list)\nplt.xlabel('Episodes')\nplt.ylabel('Total Reward')\nplt.title('Agent Performance')\nplt.show()\n <\/code><\/pre>\n\uc774 \ucf54\ub4dc\ub97c \uc2e4\ud589\ud558\uba74 \uc5d0\uc774\uc804\ud2b8\uac00 \uc5d0\ud53c\uc18c\ub4dc \ubcc4\ub85c \uc5bb\uc740 \ucd1d \ubcf4\uc0c1\uc744 \uc2dc\uac01\ud654\ud55c \uadf8\ub798\ud504\ub97c \ubcfc \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uadf8\ub798\ud504\uac00 \uc0c1\uc2b9\ud558\ub294 \uacbd\ud5a5\uc744 \ubcf4\uc778\ub2e4\uba74 \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uc5d0\uc11c \uc810\ucc28 \ub354 \ub098\uc740 \uc131\uacfc\ub97c \ub0b4\uace0 \uc788\ub2e4\ub294 \uac83\uc744 \uc758\ubbf8\ud569\ub2c8\ub2e4.<\/p>\n<\/section>\n\n7. \uace0\uae09 \uc8fc\uc81c: DQN(Deep Q-Network)<\/h2>\n
Q-\ub7ec\ub2dd\uc758 \ub2e8\uc21c\ud55c \ud615\ud0dc\ub294 \ud6a8\uacfc\uc801\uc774\uc9c0\ub9cc, \uc0c1\ud0dc \uacf5\uac04\uc774 \ud06c\uac70\ub098 \ubcf5\uc7a1\ud55c \uacbd\uc6b0\uc5d0\ub294 \uc131\ub2a5\uc5d0 \ud55c\uacc4\uac00 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \uadf9\ubcf5\ud558\uae30 \uc704\ud574 \ub525\ub7ec\ub2dd\uc744 \uacb0\ud569\ud55c DQN(Deep Q-Network) \uc54c\uace0\ub9ac\uc998\uc774 \uac1c\ubc1c\ub418\uc5c8\uc2b5\ub2c8\ub2e4. DQN\uc740 \uc2e0\uacbd\ub9dd\uc744 \uc0ac\uc6a9\ud558\uc5ec Q-\uac12\uc744 \uadfc\uc0ac\ud558\uc5ec \ub354 \ubcf5\uc7a1\ud55c \ud658\uacbd\uc5d0\uc11c\ub3c4 \ud6a8\uacfc\uc801\uc73c\ub85c \ud559\uc2b5\ud560 \uc218 \uc788\uac8c \ud569\ub2c8\ub2e4.<\/p>\n
\nimport tensorflow as tf\nfrom collections import deque\n\nclass DQNAgent:\n def __init__(self, state_size, action_size):\n self.state_size = state_size\n self.action_size = action_size\n self.memory = deque(maxlen=2000)\n self.gamma = 0.95 # discount rate\n self.epsilon = 1.0 # exploration rate\n self.epsilon_min = 0.01\n self.epsilon_decay = 0.995\n self.model = self._build_model()\n\n def _build_model(self):\n model = tf.keras.Sequential([\n tf.keras.layers.Dense(24, input_dim=self.state_size, activation='relu'),\n tf.keras.layers.Dense(24, activation='relu'),\n tf.keras.layers.Dense(self.action_size, activation='linear')\n ])\n model.compile(loss='mse', optimizer=tf.keras.optimizers.Adam(lr=0.001))\n return model\n\n def remember(self, state, action, reward, next_state, done):\n self.memory.append((state, action, reward, next_state, done))\n\n def act(self, state):\n if np.random.rand() <= self.epsilon:\n return random.randrange(self.action_size)\n act_values = self.model.predict(state)\n return np.argmax(act_values[0]) # Returns the index of the action with the maximum value\n\n def replay(self, batch_size):\n minibatch = random.sample(self.memory, batch_size)\n for state, action, reward, next_state, done in minibatch:\n target = reward\n if not done:\n target = reward + self.gamma * np.amax(self.model.predict(next_state)[0])\n target_f = self.model.predict(state)\n target_f[0][action] = target\n self.model.fit(state, target_f, epochs=1, verbose=0)\n\n# \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5 \ucf54\ub4dc \uc0dd\ub7b5\n <\/code><\/pre>\nDQN \uc5d0\uc774\uc804\ud2b8\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubcf5\uc7a1\ud55c \ud658\uacbd\uc5d0\uc11c\ub3c4 \ud559\uc2b5\uc744 \uc2dc\ub3c4\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. DQN\uc740 Q-\ub7ec\ub2dd\uc744 \ub525\ub7ec\ub2dd\uc5d0 \ud1b5\ud569\ud558\uc5ec \ub354 \ub098\uc740 \uc131\ub2a5\uc744 \ubc1c\ud718\ud558\ub294\ub370, \uc774\ub294 \uace0\ucc28\uc6d0\uc801\uc778 \uc785\ub825\uc5d0 \ub300\ud55c \uc801\uc808\ud55c \ud589\ub3d9 \uacbd\ub85c\ub97c \ud559\uc2b5\ud558\ub294 \ub370 \uc720\uc6a9\ud569\ub2c8\ub2e4.<\/p>\n<\/section>\n\n8. \uacb0\ub860<\/h2>\n
\uc774\ubc88 \ud3ec\uc2a4\ud2b8\uc5d0\uc11c\ub294 Python\uc744 \ud65c\uc6a9\ud558\uc5ec OpenAI Gym\uc5d0\uc11c \uac15\ud654\ud559\uc2b5 \ud658\uacbd\uc744 \uad6c\ucd95\ud558\uace0 \uc5d0\uc774\uc804\ud2b8\ub97c \ud559\uc2b5\uc2dc\ud0a4\ub294 \uacfc\uc815\uc744 \uc0b4\ud3b4\ubcf4\uc558\uc2b5\ub2c8\ub2e4. \uac15\ud654\ud559\uc2b5\uc740 \ub2e4\uc591\ud55c \ubd84\uc57c\uc5d0 \uc751\uc6a9\ub420 \uc218 \uc788\ub294 \ub9e4\uc6b0 \uc720\uc6a9\ud55c \uae30\uc220\uc774\uba70, OpenAI Gym\uc744 \ud1b5\ud574 \uc2e4\ud5d8\ud560 \uc218 \uc788\ub294 \ub9ce\uc740 \ud658\uacbd\uc774 \uc874\uc7ac\ud569\ub2c8\ub2e4. \uac15\ud654\ud559\uc2b5\uc758 \uc54c\uace0\ub9ac\uc998\uc740 \uacc4\uc18d\ud574\uc11c \ubc1c\uc804\ud558\uace0 \uc788\uc73c\uba70, \uc774\ub294 \uc5f0\uad6c \ubc0f \uc2e4\uc6a9\uc801\uc778 \ubb38\uc81c \ud574\uacb0\uc5d0 \ud070 \uae30\uc5ec\ub97c \ud560 \uac83\uc785\ub2c8\ub2e4.<\/p>\n
\uc5d0\uc774\uc804\ud2b8\ub97c \ud6a8\uacfc\uc801\uc73c\ub85c \ud559\uc2b5\uc2dc\ud0a4\uae30 \uc704\ud574\uc11c\ub294 \ub2e4\uc591\ud55c \ud658\uacbd\uc5d0\uc11c \uc2e4\ud5d8\uc744 \uc9c4\ud589\ud558\uace0, \ucd5c\uc801\uc758 \ud558\uc774\ud37c\ud30c\ub77c\ubbf8\ud130\ub97c \ucc3e\ub294 \uac83\uc774 \uc911\uc694\ud569\ub2c8\ub2e4. \uc55e\uc73c\ub85c\ub3c4 \ub2e4\uc591\ud55c \uc54c\uace0\ub9ac\uc998\uacfc \ud658\uacbd\uc744 \ud0d0\uc0c9\ud558\uba70 \uadf8 \uac00\ub2a5\uc131\uc744 \ud0d0\ud5d8\ud574\ubcf4\uc2dc\uae30 \ubc14\ub78d\ub2c8\ub2e4.<\/p>\n<\/section>\n<\/article>\n
<\/body><\/p>\n","protected":false},"excerpt":{"rendered":"
1. \uac15\ud654\ud559\uc2b5(RL) \uc18c\uac1c \uac15\ud654\ud559\uc2b5( Reinforcement Learning, RL)\uc740 \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uacfc \uc0c1\ud638\uc791\uc6a9\ud558\uba74\uc11c \ubcf4\uc0c1\uc744 \ucd5c\ub300\ud654\ud558\ub294 \ubc29\ud5a5\uc73c\ub85c \ud559\uc2b5\ud558\ub294 \ubc29\ubc95\uc785\ub2c8\ub2e4. \uc774 \uacfc\uc815\uc5d0\uc11c \uc5d0\uc774\uc804\ud2b8\ub294 \uc561\uc158\uc744 \uc120\ud0dd\ud558\uace0, \ud658\uacbd\uc73c\ub85c\ubd80\ud130 \ubcf4\uc0c1\uacfc \ub2e4\uc74c \uc0c1\ud0dc\ub97c \ubc1b\uace0, \uc774\ub97c \ud1b5\ud574 \uc815\ucc45(policy)\uc744 \uac1c\uc120\ud569\ub2c8\ub2e4. RL\uc744 \uc774\ud574\ud558\uae30 \uc704\ud574\uc11c\ub294 \uba87 \uac00\uc9c0 \ud575\uc2ec \uac1c\ub150\uc744 \uc775\ud600\uc57c \ud569\ub2c8\ub2e4: \ud658\uacbd(Environment): \uc5d0\uc774\uc804\ud2b8\uac00 \uc0c1\ud638\uc791\uc6a9\ud558\ub294 \ub300\uc0c1\uc785\ub2c8\ub2e4. RL \ubb38\uc81c\ub97c \uc815\uc758\ud558\ub294 \ub370 \uc911\uc694\ud55c \uc694\uc18c\uc785\ub2c8\ub2e4. \uc0c1\ud0dc(State): \uc5d0\uc774\uc804\ud2b8\uac00 \ud604\uc7ac \ud658\uacbd\uc5d0\uc11c \uad00\uce21\ud558\ub294 \uc815\ubcf4\uc785\ub2c8\ub2e4. \ud589\ub3d9(Action): … \ub354 \ubcf4\uae30 “37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95”<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[213],"tags":[],"class_list":["post-40498","post","type-post","status-publish","format-standard","hentry","category-213"],"yoast_head":"\n37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/atmokpo.com\/w\/40498\/\" \/>\n<meta property=\"og:locale\" content=\"ko_KR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\" \/>\n<meta property=\"og:description\" content=\"1. \uac15\ud654\ud559\uc2b5(RL) \uc18c\uac1c \uac15\ud654\ud559\uc2b5( Reinforcement Learning, RL)\uc740 \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uacfc \uc0c1\ud638\uc791\uc6a9\ud558\uba74\uc11c \ubcf4\uc0c1\uc744 \ucd5c\ub300\ud654\ud558\ub294 \ubc29\ud5a5\uc73c\ub85c \ud559\uc2b5\ud558\ub294 \ubc29\ubc95\uc785\ub2c8\ub2e4. \uc774 \uacfc\uc815\uc5d0\uc11c \uc5d0\uc774\uc804\ud2b8\ub294 \uc561\uc158\uc744 \uc120\ud0dd\ud558\uace0, \ud658\uacbd\uc73c\ub85c\ubd80\ud130 \ubcf4\uc0c1\uacfc \ub2e4\uc74c \uc0c1\ud0dc\ub97c \ubc1b\uace0, \uc774\ub97c \ud1b5\ud574 \uc815\ucc45(policy)\uc744 \uac1c\uc120\ud569\ub2c8\ub2e4. RL\uc744 \uc774\ud574\ud558\uae30 \uc704\ud574\uc11c\ub294 \uba87 \uac00\uc9c0 \ud575\uc2ec \uac1c\ub150\uc744 \uc775\ud600\uc57c \ud569\ub2c8\ub2e4: \ud658\uacbd(Environment): \uc5d0\uc774\uc804\ud2b8\uac00 \uc0c1\ud638\uc791\uc6a9\ud558\ub294 \ub300\uc0c1\uc785\ub2c8\ub2e4. RL \ubb38\uc81c\ub97c \uc815\uc758\ud558\ub294 \ub370 \uc911\uc694\ud55c \uc694\uc18c\uc785\ub2c8\ub2e4. \uc0c1\ud0dc(State): \uc5d0\uc774\uc804\ud2b8\uac00 \ud604\uc7ac \ud658\uacbd\uc5d0\uc11c \uad00\uce21\ud558\ub294 \uc815\ubcf4\uc785\ub2c8\ub2e4. \ud589\ub3d9(Action): … \ub354 \ubcf4\uae30 "37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95"\" \/>\n<meta property=\"og:url\" content=\"https:\/\/atmokpo.com\/w\/40498\/\" \/>\n<meta property=\"og:site_name\" content=\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\" \/>\n<meta property=\"article:published_time\" content=\"2024-11-04T01:52:13+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2024-11-26T06:39:01+00:00\" \/>\n<meta name=\"author\" content=\"root\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@bebubo4\" \/>\n<meta name=\"twitter:site\" content=\"@bebubo4\" \/>\n<meta name=\"twitter:label1\" content=\"\uae00\uc4f4\uc774\" \/>\n\t<meta name=\"twitter:data1\" content=\"root\" \/>\n\t<meta name=\"twitter:label2\" content=\"\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04\" \/>\n\t<meta name=\"twitter:data2\" content=\"3\ubd84\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/\"},\"author\":{\"name\":\"root\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7\"},\"headline\":\"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95\",\"datePublished\":\"2024-11-04T01:52:13+00:00\",\"dateModified\":\"2024-11-26T06:39:01+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/\"},\"wordCount\":78,\"publisher\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\"},\"articleSection\":[\"\uac15\ud654\ud559\uc2b5\"],\"inLanguage\":\"ko-KR\"},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/\",\"url\":\"https:\/\/atmokpo.com\/w\/40498\/\",\"name\":\"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"isPartOf\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#website\"},\"datePublished\":\"2024-11-04T01:52:13+00:00\",\"dateModified\":\"2024-11-26T06:39:01+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/#breadcrumb\"},\"inLanguage\":\"ko-KR\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/atmokpo.com\/w\/40498\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/atmokpo.com\/w\/40498\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"\ud648\",\"item\":\"https:\/\/atmokpo.com\/w\/en\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/atmokpo.com\/w\/#website\",\"url\":\"https:\/\/atmokpo.com\/w\/\",\"name\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/atmokpo.com\/w\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"ko-KR\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/atmokpo.com\/w\/#organization\",\"name\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\",\"url\":\"https:\/\/atmokpo.com\/w\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png\",\"contentUrl\":\"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png\",\"width\":400,\"height\":400,\"caption\":\"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8\"},\"image\":{\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/x.com\/bebubo4\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7\",\"name\":\"root\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"ko-KR\",\"@id\":\"https:\/\/atmokpo.com\/w\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g\",\"caption\":\"root\"},\"sameAs\":[\"http:\/\/atmokpo.com\/w\"],\"url\":\"https:\/\/atmokpo.com\/w\/author\/root\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/atmokpo.com\/w\/40498\/","og_locale":"ko_KR","og_type":"article","og_title":"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","og_description":"1. \uac15\ud654\ud559\uc2b5(RL) \uc18c\uac1c \uac15\ud654\ud559\uc2b5( Reinforcement Learning, RL)\uc740 \uc5d0\uc774\uc804\ud2b8\uac00 \ud658\uacbd\uacfc \uc0c1\ud638\uc791\uc6a9\ud558\uba74\uc11c \ubcf4\uc0c1\uc744 \ucd5c\ub300\ud654\ud558\ub294 \ubc29\ud5a5\uc73c\ub85c \ud559\uc2b5\ud558\ub294 \ubc29\ubc95\uc785\ub2c8\ub2e4. \uc774 \uacfc\uc815\uc5d0\uc11c \uc5d0\uc774\uc804\ud2b8\ub294 \uc561\uc158\uc744 \uc120\ud0dd\ud558\uace0, \ud658\uacbd\uc73c\ub85c\ubd80\ud130 \ubcf4\uc0c1\uacfc \ub2e4\uc74c \uc0c1\ud0dc\ub97c \ubc1b\uace0, \uc774\ub97c \ud1b5\ud574 \uc815\ucc45(policy)\uc744 \uac1c\uc120\ud569\ub2c8\ub2e4. RL\uc744 \uc774\ud574\ud558\uae30 \uc704\ud574\uc11c\ub294 \uba87 \uac00\uc9c0 \ud575\uc2ec \uac1c\ub150\uc744 \uc775\ud600\uc57c \ud569\ub2c8\ub2e4: \ud658\uacbd(Environment): \uc5d0\uc774\uc804\ud2b8\uac00 \uc0c1\ud638\uc791\uc6a9\ud558\ub294 \ub300\uc0c1\uc785\ub2c8\ub2e4. RL \ubb38\uc81c\ub97c \uc815\uc758\ud558\ub294 \ub370 \uc911\uc694\ud55c \uc694\uc18c\uc785\ub2c8\ub2e4. \uc0c1\ud0dc(State): \uc5d0\uc774\uc804\ud2b8\uac00 \ud604\uc7ac \ud658\uacbd\uc5d0\uc11c \uad00\uce21\ud558\ub294 \uc815\ubcf4\uc785\ub2c8\ub2e4. \ud589\ub3d9(Action): … \ub354 \ubcf4\uae30 \"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95\"","og_url":"https:\/\/atmokpo.com\/w\/40498\/","og_site_name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","article_published_time":"2024-11-04T01:52:13+00:00","article_modified_time":"2024-11-26T06:39:01+00:00","author":"root","twitter_card":"summary_large_image","twitter_creator":"@bebubo4","twitter_site":"@bebubo4","twitter_misc":{"\uae00\uc4f4\uc774":"root","\uc608\uc0c1 \ub418\ub294 \ud310\ub3c5 \uc2dc\uac04":"3\ubd84"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/atmokpo.com\/w\/40498\/#article","isPartOf":{"@id":"https:\/\/atmokpo.com\/w\/40498\/"},"author":{"name":"root","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7"},"headline":"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95","datePublished":"2024-11-04T01:52:13+00:00","dateModified":"2024-11-26T06:39:01+00:00","mainEntityOfPage":{"@id":"https:\/\/atmokpo.com\/w\/40498\/"},"wordCount":78,"publisher":{"@id":"https:\/\/atmokpo.com\/w\/#organization"},"articleSection":["\uac15\ud654\ud559\uc2b5"],"inLanguage":"ko-KR"},{"@type":"WebPage","@id":"https:\/\/atmokpo.com\/w\/40498\/","url":"https:\/\/atmokpo.com\/w\/40498\/","name":"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95 - \ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","isPartOf":{"@id":"https:\/\/atmokpo.com\/w\/#website"},"datePublished":"2024-11-04T01:52:13+00:00","dateModified":"2024-11-26T06:39:01+00:00","breadcrumb":{"@id":"https:\/\/atmokpo.com\/w\/40498\/#breadcrumb"},"inLanguage":"ko-KR","potentialAction":[{"@type":"ReadAction","target":["https:\/\/atmokpo.com\/w\/40498\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/atmokpo.com\/w\/40498\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"\ud648","item":"https:\/\/atmokpo.com\/w\/en\/"},{"@type":"ListItem","position":2,"name":"37. Python\uc744 \ud65c\uc6a9\ud55c \uac15\ud654\ud559\uc2b5 \ud658\uacbd \uad6c\ucd95 \ubc0f \uc5d0\uc774\uc804\ud2b8 \ud559\uc2b5, OpenAI Gym \ub4f1 RL \ud658\uacbd \uc124\uc815 \ubc29\ubc95"}]},{"@type":"WebSite","@id":"https:\/\/atmokpo.com\/w\/#website","url":"https:\/\/atmokpo.com\/w\/","name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","description":"","publisher":{"@id":"https:\/\/atmokpo.com\/w\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/atmokpo.com\/w\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"ko-KR"},{"@type":"Organization","@id":"https:\/\/atmokpo.com\/w\/#organization","name":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8","url":"https:\/\/atmokpo.com\/w\/","logo":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/","url":"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png","contentUrl":"https:\/\/atmokpo.com\/w\/wp-content\/uploads\/2024\/11\/logo.png","width":400,"height":400,"caption":"\ub77c\uc774\ube0c\uc2a4\ub9c8\ud2b8"},"image":{"@id":"https:\/\/atmokpo.com\/w\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/bebubo4"]},{"@type":"Person","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/91b6b3b138fbba0efb4ae64b1abd81d7","name":"root","image":{"@type":"ImageObject","inLanguage":"ko-KR","@id":"https:\/\/atmokpo.com\/w\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/708197b41fc6435a7ce22d951b25d4a47e9e904270cb1f04682d4f025066f80c?s=96&d=mm&r=g","caption":"root"},"sameAs":["http:\/\/atmokpo.com\/w"],"url":"https:\/\/atmokpo.com\/w\/author\/root\/"}]}},"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"jetpack-related-posts":[],"_links":{"self":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/40498","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/comments?post=40498"}],"version-history":[{"count":1,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/40498\/revisions"}],"predecessor-version":[{"id":40499,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/posts\/40498\/revisions\/40499"}],"wp:attachment":[{"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/media?parent=40498"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/categories?post=40498"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/atmokpo.com\/w\/wp-json\/wp\/v2\/tags?post=40498"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}