\ubcf8 \uae00\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uacfc \ud30c\uc774\ud1a0\uce58\ub97c \ud65c\uc6a9\ud558\uc5ec \ubaac\ud14c\uce74\ub97c\ub85c \ud2b8\ub9ac \uac80\uc0c9(MCTS) \uc54c\uace0\ub9ac\uc998\uc744 \uc801\uc6a9\ud55c \ud2f1\ud0dd\ud1a0 \uac8c\uc784\uc744 \uad6c\ud604\ud558\ub294 \uacfc\uc815\uc744 \uc124\uba85\ud569\ub2c8\ub2e4. \uc6b0\ub9ac\ub294 \uae30\ubcf8\uc801\uc73c\ub85c MCTS\uac00 \uc5b4\ub5bb\uac8c \uc791\ub3d9\ud558\ub294\uc9c0, \uadf8\ub9ac\uace0 \uc774\ub97c \ud1b5\ud574 \uc5b4\ub5bb\uac8c AI\uac00 \ud2f1\ud0dd\ud1a0 \uac8c\uc784\uc744 \ud50c\ub808\uc774\ud560 \uc218 \uc788\ub294\uc9c0\ub97c \uc774\ud574\ud560 \uac83\uc785\ub2c8\ub2e4.<\/p>\n
\ud2f1\ud0dd\ud1a0 \uac8c\uc784 \uac1c\uc694<\/h2>\n
\ud2f1\ud0dd\ud1a0(Tic-Tac-Toe)\ub294 3×3 \uc815\uc0ac\uac01\ud615 \uaca9\uc790\uc5d0\uc11c \ub450 \ud50c\ub808\uc774\uc5b4\uac00 \ubc88\uac08\uc544 \uac00\uba70 X \ub610\ub294 O\ub97c \ub193\uace0, \uac00\ub85c, \uc138\ub85c \ub610\ub294 \ub300\uac01\uc120\uc73c\ub85c 3\uac1c\uc758 \uc5f0\uc18d\ub41c \ub9d0\uc744 \ub193\uc73c\uba74 \uc2b9\ub9ac\ud558\ub294 \uac8c\uc784\uc785\ub2c8\ub2e4.<\/p>\n
1\ub2e8\uacc4: \ud658\uacbd \uc124\uc815<\/h2>\n
\uc774 \ud29c\ud1a0\ub9ac\uc5bc\uc744 \uc9c4\ud589\ud558\uae30 \uc704\ud574\uc11c\ub294 \ud544\uc694\ud55c \ud328\ud0a4\uc9c0\ub97c \uc124\uce58\ud574\uc57c \ud569\ub2c8\ub2e4. \ub2e4\uc74c\uc740 \ud544\uc694\ud55c \uc8fc\uc694 \ub77c\uc774\ube0c\ub7ec\ub9ac\uc785\ub2c8\ub2e4.<\/p>\n
pip install torch numpy matplotlib<\/code><\/pre>\n2\ub2e8\uacc4: \uac8c\uc784 \ud658\uacbd \uad6c\ud604<\/h2>\n
\uba3c\uc800 \ud2f1\ud0dd\ud1a0 \uac8c\uc784 \ud658\uacbd\uc744 \uad6c\ud604\ud569\ub2c8\ub2e4. \uac8c\uc784\uc758 \uaddc\uce59\uc744 \uc815\uc758\ud558\uace0, \uc0c1\ud0dc\ub97c \ub098\ud0c0\ub0b4\ub294 \ud074\ub798\uc2a4\ub97c \ub9cc\ub4e4\uc5b4\uc57c \ud569\ub2c8\ub2e4.<\/p>\n
\nimport numpy as np\n\nclass TicTacToe:\n def __init__(self):\n self.board = np.zeros((3, 3), dtype=int) # 0: \ube48\uce78, 1: X, -1: O\n self.current_player = 1 # 1: X\uc758 \ucc28\ub840, -1: O\uc758 \ucc28\ub840\n\n def reset(self):\n self.board = np.zeros((3, 3), dtype=int)\n self.current_player = 1\n\n def make_move(self, row, col):\n if self.board[row, col] == 0:\n self.board[row, col] = self.current_player\n self.current_player *= -1\n\n def check_winner(self):\n for player in [1, -1]:\n for row in range(3):\n if np.all(self.board[row, :] == player): # \ud589 \uccb4\ud06c\n return player\n for col in range(3):\n if np.all(self.board[:, col] == player): # \uc5f4 \uccb4\ud06c\n return player\n if np.all(np.diag(self.board) == player) or np.all(np.diag(np.fliplr(self.board)) == player):\n return player\n return None if np.any(self.board == 0) else 0 # \uac8c\uc784\uc774 \uc9c4\ud589 \uc911\uc778 \uacbd\uc6b0\n \n def display(self):\n symbols = {1: 'X', -1: 'O', 0: ' '}\n for row in self.board:\n print(\"|\".join(symbols[x] for x in row))\n print(\"-\" * 5)\n print(\"\\n\")\n\n# \uac8c\uc784 \ud14c\uc2a4\ud2b8\ngame = TicTacToe()\ngame.make_move(0, 0)\ngame.display()\ngame.make_move(1, 1)\ngame.display()\n <\/code><\/pre>\n3\ub2e8\uacc4: \ubaac\ud14c\uce74\ub97c\ub85c \ud2b8\ub9ac \uac80\uc0c9(MCTS) \uc54c\uace0\ub9ac\uc998<\/h2>\n
MCTS\ub294 \ubd88\ud655\uc2e4\ud55c \uc0c1\ud669\uc5d0\uc11c \uc758\uc0ac\uacb0\uc815 \ubb38\uc81c\ub97c \ud574\uacb0\ud558\uae30 \uc704\ud55c \ubc29\ubc95\uc785\ub2c8\ub2e4. \uae30\ubcf8\uc801\uc73c\ub85c \uc774 \uc54c\uace0\ub9ac\uc998\uc740 \ub2e4\uc74c \ub124 \uac00\uc9c0 \ub2e8\uacc4\ub85c \uad6c\uc131\ub429\ub2c8\ub2e4:<\/p>\n
\n- \uc120\ud0dd(Selection)<\/strong>: \ud604\uc7ac \ud2b8\ub9ac\uc5d0\uc11c \ub178\ub4dc\ub97c \uc120\ud0dd\ud569\ub2c8\ub2e4.<\/li>\n
- \ud655\uc7a5(Expansion)<\/strong>: \uc120\ud0dd\ub41c \ub178\ub4dc\uc5d0\uc11c \uac00\ub2a5\ud55c \ud589\ub3d9\uc744 \ud655\uc7a5\ud569\ub2c8\ub2e4.<\/li>\n
- \uc2dc\ubbac\ub808\uc774\uc158(Simulation)<\/strong>: \ud655\uc7a5\ub41c \ub178\ub4dc\uc5d0\uc11c \uac8c\uc784\uc744 \ud50c\ub808\uc774\ud558\uc5ec \uacb0\uacfc\ub97c \uc5bb\uc2b5\ub2c8\ub2e4.<\/li>\n
- \ubc31\uc5c5(Backpropagation)<\/strong>: \uacb0\uacfc\ub97c \ud1b5\ud574 \ubd80\ubaa8 \ub178\ub4dc\uc5d0 \uc815\ubcf4\ub97c \uc5c5\ub370\uc774\ud2b8\ud569\ub2c8\ub2e4.<\/li>\n<\/ol>\n
MCTS \ud074\ub798\uc2a4 \uad6c\ud604<\/h3>\n\nimport random\nfrom collections import defaultdict\n\nclass MCTSNode:\n def __init__(self, state, parent=None):\n self.state = state # \ud604\uc7ac \uac8c\uc784 \uc0c1\ud0dc\n self.parent = parent\n self.children = [] # \uc790\uc2dd \ub178\ub4dc\n self.wins = 0 # \uc2b9\ub9ac \ud69f\uc218\n self.visits = 0 # \ubc29\ubb38 \ud69f\uc218\n\n def ucb1(self):\n if self.visits == 0:\n return float('inf') # \ubc29\ubb38\ud55c \uc801 \uc5c6\ub294 \ub178\ub4dc\ub294 \uc6b0\uc120 \uc120\ud0dd\ud558\ub3c4\ub85d \ud568\n return self.wins \/ self.visits + np.sqrt(2 * np.log(self.parent.visits) \/ self.visits)\n\nclass MCTS:\n def __init__(self, iterations):\n self.iterations = iterations\n\n def search(self, game):\n root = MCTSNode(state=game)\n\n for _ in range(self.iterations):\n node = self.select(root)\n winner = self.simulate(node.state)\n self.backpropagate(node, winner)\n\n return max(root.children, key=lambda child: child.visits).state\n\n def select(self, node):\n while node.children:\n node = max(node.children, key=lambda child: child.ucb1())\n if node.visits > 0:\n for action in self.get_valid_moves(node.state):\n child_state = node.state.copy()\n child_state.make_move(action[0], action[1])\n child_node = MCTSNode(state=child_state, parent=node)\n node.children.append(child_node)\n return random.choice(node.children) if node.children else node\n\n def simulate(self, state):\n current_player = state.current_player\n while True:\n winner = state.check_winner()\n if winner is not None:\n return winner\n valid_moves = self.get_valid_moves(state)\n move = random.choice(valid_moves)\n state.make_move(move[0], move[1])\n\n def backpropagate(self, node, winner):\n while node is not None:\n node.visits += 1\n if winner == 1: # X\uc758 \uc2b9\ub9ac\n node.wins += 1\n node = node.parent\n\n def get_valid_moves(self, state):\n return [(row, col) for row in range(3) for col in range(3) if state.board[row, col] == 0]\n\n# MCTS \uc0ac\uc6a9 \uc0ac\ub840\nmcts = MCTS(iterations=1000)\nmove = mcts.search(game)\nprint(\"AI\uc758 \uc120\ud0dd:\", move)\n <\/code><\/pre>\n4\ub2e8\uacc4: AI\uc640 \uc0ac\uc6a9\uc790 \uac04\uc758 \uac8c\uc784 \uad6c\ud604<\/h2>\n
\uc774\uc81c \uc644\uc131\ub41c MCTS\ub97c \uc0ac\uc6a9\ud558\uc5ec \uc0ac\uc6a9\uc790\uc640 AI \uac04\uc758 \uac8c\uc784\uc744 \uad6c\ud604\ud574 \ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n
\ndef play_game():\n game = TicTacToe()\n while True:\n game.display()\n if game.current_player == 1: # \uc0ac\uc6a9\uc790 \ucc28\ub840\n row, col = map(int, input(\"\ud589\uacfc \uc5f4 \ubc88\ud638\ub97c \uc785\ub825\ud558\uc138\uc694 (0, 1 \ub610\ub294 2): \").split())\n game.make_move(row, col)\n else: # AI \ucc28\ub840\n print(\"AI\uac00 \uc120\ud0dd\ud558\ub294 \uc911...\")\n move = mcts.search(game)\n game.make_move(move[0], move[1])\n print(f\"AI\uac00 \uc120\ud0dd\ud55c \uc704\uce58: {move}\")\n\n winner = game.check_winner()\n if winner is not None:\n game.display()\n if winner == 1:\n print(\"\ucd95\ud558\ud569\ub2c8\ub2e4! \ub2f9\uc2e0\uc774 \uc774\uacbc\uc2b5\ub2c8\ub2e4!\")\n elif winner == -1:\n print(\"AI\uac00 \uc774\uacbc\uc2b5\ub2c8\ub2e4!\")\n else:\n print(\"\ubb34\uc2b9\ubd80\uc785\ub2c8\ub2e4!\")\n break\n\nplay_game()\n <\/code><\/pre>\n\uacb0\ub860<\/h2>\n
\uc774\ubc88 \uac15\uc88c\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uacfc \ud30c\uc774\ud1a0\uce58\uc758 \uae30\ubcf8 \uac00\uc774\ub4dc\ub77c\uc778\uc744 \uc0b4\ud3b4\ubcf4\uc558\uc2b5\ub2c8\ub2e4. \ubaac\ud14c\uce74\ub97c\ub85c \ud2b8\ub9ac \uac80\uc0c9\uc744 \ud65c\uc6a9\ud558\uc5ec \uac04\ub2e8\ud55c \ud2f1\ud0dd\ud1a0 AI\ub97c \uad6c\ud604\ud558\ub294 \uacfc\uc815\uc740 \uae30\uc220\uc801\uc73c\ub85c \ub3c4\uc804\uc774 \ub420 \uc218 \uc788\uc9c0\ub9cc, \uacb0\uacfc\uc801\uc73c\ub85c\ub294 \ub9e4\uc6b0 \ud765\ubbf8\ub85c\uc6b4 \uacbd\ud5d8\uc774\uc5c8\uc2b5\ub2c8\ub2e4. \uc55e\uc73c\ub85c \ub354 \ub098\uc544\uac00 \ub2e4\uc591\ud55c \uc54c\uace0\ub9ac\uc998\uacfc \uae30\uc220\uc744 \ud65c\uc6a9\ud558\uc5ec \ubcf4\ub2e4 \uc644\ubcbd\ud55c AI\ub97c \uac1c\ubc1c\ud560 \uc218 \uc788\uae30\ub97c \uae30\ub300\ud569\ub2c8\ub2e4.<\/p>\n<\/article>\n
<\/body><\/p>\n","protected":false},"excerpt":{"rendered":"