{"id":1116,"date":"2026-03-16T15:41:54","date_gmt":"2026-03-16T07:41:54","guid":{"rendered":"https:\/\/www.ndnlab.com\/?p=1116"},"modified":"2026-03-16T15:41:55","modified_gmt":"2026-03-16T07:41:55","slug":"liquid-graph-time-constant-networkfor-multi-agent-systems-control","status":"publish","type":"post","link":"https:\/\/www.ndnlab.com\/?p=1116","title":{"rendered":"Liquid-Graph Time-Constant Networkfor Multi-Agent Systems Control"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">\u2460\u7814\u7a76\u80cc\u666f\uff1a<\/h2>\n\n\n\n<p>\u5728\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\uff08Multi-Agent Systems, MAS\uff09\u4e2d\uff0c\u901a\u4fe1\u662f\u5b9e\u73b0\u5206\u5e03\u5f0f\u63a7\u5236\u548c\u89c4\u5212\u7684\u5173\u952e\u8981\u7d20\u3002\u901a\u8fc7\u667a\u80fd\u4f53\u4e4b\u95f4\u7684\u4fe1\u606f\u4ea4\u6362\uff0c\u53ef\u4ee5\u5c06\u9884\u6d4b\u548c\u51b3\u7b56\u8fc7\u7a0b\u5206\u5e03\u5230\u591a\u4e2a\u8282\u70b9\u4e0a\uff0c\u4ece\u800c\u63d0\u9ad8\u7cfb\u7edf\u7684\u53ef\u6269\u5c55\u6027\u4e0e\u9884\u6d4b\u80fd\u529b\u3002<br>\u8fd1\u5e74\u6765\uff0c\u968f\u7740\u6570\u636e\u9a71\u52a8\u63a7\u5236\u65b9\u6cd5\u7684\u53d1\u5c55\uff0c\u8d8a\u6765\u8d8a\u591a\u7814\u7a76\u5c1d\u8bd5\u5229\u7528 <strong>\u56fe\u795e\u7ecf\u7f51\u7edc\uff08Graph Neural Networks, GNNs\uff09<\/strong> \u6765\u5efa\u6a21\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u4e2d\u7684\u4fe1\u606f\u4f20\u64ad\u4e0e\u51b3\u7b56\u8fc7\u7a0b\u3002\u4f8b\u5982\u5df2\u6709\u5de5\u4f5c\u5c06 GNN \u5e94\u7528\u4e8e <strong>\u7fa4\u4f53\u7f16\u961f\u63a7\u5236\uff08flocking\uff09\u3001\u7a7a\u95f4\u8986\u76d6\u3001\u591a\u673a\u5668\u4eba\u8def\u5f84\u89c4\u5212\u4ee5\u53ca\u8fd0\u52a8\u89c4\u5212\u7b49\u4efb\u52a1<\/strong>\uff0c\u5e76\u53d6\u5f97\u4e86\u4e00\u5b9a\u6548\u679c\u3002\u4e0e\u6b64\u540c\u65f6\uff0c\u7814\u7a76\u8005\u4e5f\u9010\u6e10\u5173\u6ce8<strong>\u5b66\u4e60\u63a7\u5236\u65b9\u6cd5\u7684\u7a33\u5b9a\u6027\u95ee\u9898<\/strong>\u3002\u5df2\u6709\u7814\u7a76\u5c1d\u8bd5\u901a\u8fc7<strong>\u6536\u7f29\u5206\u6790\uff08contraction analysis\uff09<\/strong>\u6765\u8bc1\u660e\u9012\u5f52\u795e\u7ecf\u7f51\u7edc\u6216\u63a7\u5236\u7cfb\u7edf\u7684\u7a33\u5b9a\u6027\uff0c\u5e76\u5728 LSTM\u3001GRU \u7b49\u6a21\u578b\u4e2d\u5f15\u5165<strong>\u8f93\u5165\u5230\u72b6\u6001\u7a33\u5b9a\u6027\uff08ISS\uff09\u548c\u589e\u91cf IS\uff08\u03b4ISS\uff09<\/strong> \u7684\u7406\u8bba\u6846\u67b6\uff0c\u4ee5\u4fdd\u8bc1\u5b66\u4e60\u63a7\u5236\u7cfb\u7edf\u5728\u52a8\u6001\u73af\u5883\u4e2d\u7684\u7a33\u5b9a\u6027\u3002\u5728\u8fd9\u6837\u7684\u7814\u7a76\u80cc\u666f\u4e0b\uff0c\u4e00\u4e2a\u65b0\u7684\u95ee\u9898\u9010\u6e10\u51fa\u73b0\uff1a\u5982\u4f55\u8bbe\u8ba1\u4e00\u79cd\u65e2\u80fd\u5229\u7528 <strong>\u56fe\u7ed3\u6784\u4fe1\u606f\u4f20\u64ad\u80fd\u529b<\/strong>\uff0c\u53c8\u80fd\u523b\u753b <strong>\u8fde\u7eed\u65f6\u95f4\u52a8\u529b\u5b66\u884c\u4e3a<\/strong> \u7684\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\uff0c\u7528\u4e8e\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u7684\u5206\u5e03\u5f0f\u63a7\u5236\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2461\u4f20\u7edf\u65b9\u6cd5<\/h2>\n\n\n\n<p>\u5c3d\u7ba1\u56fe\u795e\u7ecf\u7f51\u7edc\u5728\u591a\u667a\u80fd\u4f53\u63a7\u5236\u4e2d\u8868\u73b0\u51fa\u826f\u597d\u7684\u5efa\u6a21\u80fd\u529b\uff0c\u4f46\u73b0\u6709\u65b9\u6cd5\u4ecd\u7136\u5b58\u5728\u4e00\u4e9b\u5c40\u9650\u3002\u9996\u5148\uff0c\u5927\u591a\u6570 GNN \u6a21\u578b\u672c\u8d28\u4e0a\u662f <strong>\u79bb\u6563\u65f6\u95f4\u6a21\u578b<\/strong>\uff0c\u5b83\u4eec\u901a\u8fc7\u5c42\u7ea7\u4f20\u64ad\u6216\u9012\u5f52\u7ed3\u6784\u8fdb\u884c\u72b6\u6001\u66f4\u65b0\uff0c\u800c\u771f\u5b9e\u7269\u7406\u7cfb\u7edf\u5f80\u5f80\u5177\u6709 <strong>\u8fde\u7eed\u65f6\u95f4\u52a8\u529b\u5b66\u7279\u5f81<\/strong>\u3002\u56e0\u6b64\uff0c\u79bb\u6563\u6a21\u578b\u5728\u523b\u753b\u590d\u6742\u52a8\u6001\u7cfb\u7edf\u65f6\u53ef\u80fd\u5b58\u5728\u8868\u8fbe\u80fd\u529b\u4e0d\u8db3\u7684\u95ee\u9898\u3002\u5176\u6b21\uff0c\u4f20\u7edf\u56fe\u795e\u7ecf\u7f51\u7edc\u5728\u591a\u667a\u80fd\u4f53\u573a\u666f\u4e2d\u901a\u5e38\u9700\u8981 <strong>\u5927\u91cf\u901a\u4fe1\u53d8\u91cf<\/strong>\u3002\u6bcf\u4e2a\u8282\u70b9\u5728\u6bcf\u4e00\u5c42\u7f51\u7edc\u4f20\u64ad\u4e2d\u90fd\u9700\u8981\u4ea4\u6362\u9690\u85cf\u72b6\u6001\u6216\u7279\u5f81\u5411\u91cf\uff0c\u8fd9\u4f1a\u5e26\u6765\u8f83\u9ad8\u7684\u901a\u4fe1\u8d1f\u62c5\uff0c\u5c24\u5176\u662f\u5728\u5927\u89c4\u6a21\u667a\u80fd\u4f53\u7cfb\u7edf\u4e2d\u3002\u53e6\u5916\uff0c\u4e00\u4e9b\u56fe\u9012\u5f52\u795e\u7ecf\u7f51\u7edc\uff08Graph Recurrent Neural Networks\uff09\u5728\u5904\u7406\u957f\u5e8f\u5217\u4fe1\u606f\u65f6\u4f1a\u9047\u5230 <strong>\u68af\u5ea6\u6d88\u5931\u95ee\u9898\u4ee5\u53ca\u4fe1\u606f\u4f20\u64ad\u4e0d\u5e73\u8861\u7684\u95ee\u9898<\/strong>\uff0c\u4f8b\u5982\u56fe\u4e2d\u67d0\u4e9b\u8def\u5f84\u6216\u8282\u70b9\u53ef\u80fd\u5728\u957f\u671f\u4fe1\u606f\u4f20\u64ad\u8fc7\u7a0b\u4e2d\u88ab\u8fc7\u5ea6\u5f3a\u5316\u6216\u5ffd\u7565\u3002\u56e0\u6b64\uff0c\u4e00\u4e2a\u7406\u60f3\u7684\u6a21\u578b\u9700\u8981\u540c\u65f6\u6ee1\u8db3\u51e0\u4e2a\u6761\u4ef6\uff1a\u4e00\u65b9\u9762\u80fd\u591f\u63cf\u8ff0 <strong>\u8fde\u7eed\u65f6\u95f4\u52a8\u6001\u7cfb\u7edf<\/strong>\uff0c\u53e6\u4e00\u65b9\u9762\u8981\u5728 <strong>\u4fdd\u6301\u8868\u8fbe\u80fd\u529b\u7684\u540c\u65f6\u964d\u4f4e\u901a\u4fe1\u5f00\u9500<\/strong>\uff0c\u5e76\u4e14\u5728\u7406\u8bba\u4e0a\u80fd\u591f\u4fdd\u8bc1\u7cfb\u7edf\u7a33\u5b9a\u6027\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2462LGTC\u6a21\u578b<\/h2>\n\n\n\n<p>\u9488\u5bf9\u4e0a\u8ff0\u95ee\u9898\uff0c\u8bba\u6587\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u6a21\u578b\uff1a<strong>Liquid-Graph Time-Constant (LGTC) Network<\/strong>\u3002LGTC \u53ef\u4ee5\u770b\u4f5c\u662f\u4e00\u79cd <strong>\u8fde\u7eed\u65f6\u95f4\u56fe\u795e\u7ecf\u7f51\u7edc\u6a21\u578b<\/strong>\uff0c\u5b83\u53d7\u5230 <strong>Liquid Time-Constant (LTC) \u7f51\u7edc<\/strong> \u7684\u542f\u53d1\u3002LTC \u7f51\u7edc\u662f\u4e00\u79cd\u57fa\u4e8e <strong>\u5e38\u5fae\u5206\u65b9\u7a0b\uff08ODE\uff09\u63cf\u8ff0\u7684\u8fde\u7eed\u65f6\u95f4\u795e\u7ecf\u7f51\u7edc<\/strong>\uff0c\u5176\u65f6\u95f4\u5e38\u6570\u662f\u52a8\u6001\u53d8\u5316\u7684\uff0c\u53ef\u4ee5\u6839\u636e\u8f93\u5165\u548c\u5185\u90e8\u72b6\u6001\u8fdb\u884c\u8c03\u6574\uff0c\u56e0\u6b64\u5177\u6709\u8f83\u5f3a\u7684\u8868\u793a\u80fd\u529b\u3002<\/p>\n\n\n\n<p>\u5728 LGTC \u4e2d\uff0c\u6bcf\u4e2a\u8282\u70b9\u7684\u72b6\u6001\u6f14\u5316\u7531\u4ee5\u4e0b\u5fae\u5206\u65b9\u7a0b\u63cf\u8ff0\uff1a<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"671\" height=\"113\" src=\"https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-13.png\"  class=\"wp-image-1118\" srcset=\"https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-13.png 671w, https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-13-300x51.png 300w\" sizes=\"auto, (max-width: 671px) 100vw, 671px\" title=\"Liquid-Graph Time-Constant Networkfor Multi-Agent Systems Control\u63d2\u56fe\" alt=\"Liquid-Graph Time-Constant Networkfor Multi-Agent Systems Control\u63d2\u56fe\" \/><\/figure>\n\n\n\n<p>\u5176\u4e2d\u8282\u70b9\u4e4b\u95f4\u7684\u4ea4\u4e92\u7531\u56fe\u7ed3\u6784\u4f20\u64ad\u7b97\u5b50 <math><semantics><mrow><mi>S<\/mi><\/mrow><annotation encoding=\"application\/x-tex\">S<\/annotation><\/semantics><\/math>S \u8868\u793a\uff0c\u4f8b\u5982 <strong>\u62c9\u666e\u62c9\u65af\u77e9\u9635\u6216\u90bb\u63a5\u77e9\u9635<\/strong>\u3002\u8fd9\u79cd\u7ed3\u6784\u4f7f\u5f97\u7f51\u7edc\u65e2\u80fd\u523b\u753b <strong>\u8282\u70b9\u4e4b\u95f4\u7684\u56fe\u7ed3\u6784\u5173\u7cfb<\/strong>\uff0c\u53c8\u80fd\u8868\u793a <strong>\u8fde\u7eed\u65f6\u95f4\u52a8\u6001\u884c\u4e3a<\/strong>\u3002LGTC \u7684\u4e00\u4e2a\u91cd\u8981\u7279\u70b9\u662f\u5f15\u5165\u4e86 <strong>\u52a8\u6001\u65f6\u95f4\u5e38\u6570\uff08dynamic time constant\uff09<\/strong>\u3002\u65f6\u95f4\u5e38\u6570\u4e0d\u4ec5\u4f9d\u8d56\u8f93\u5165\u6570\u636e\uff0c\u8fd8\u4f9d\u8d56\u4e8e\u8282\u70b9\u5185\u90e8\u72b6\u6001\u4ee5\u53ca\u6765\u81ea\u90bb\u5c45\u8282\u70b9\u7684\u901a\u4fe1\u4fe1\u606f\u3002\u8fd9\u6837\uff0c\u6bcf\u4e2a\u9690\u85cf\u72b6\u6001\u7ef4\u5ea6\u90fd\u53ef\u4ee5\u6355\u83b7\u4e0d\u540c\u7684\u52a8\u6001\u6a21\u5f0f\uff0c\u4ece\u800c\u63d0\u5347\u6a21\u578b\u7684\u9884\u6d4b\u80fd\u529b\u3002\u6b64\u5916\uff0c\u8bba\u6587\u8fd8\u5229\u7528 <strong>\u6536\u7f29\u7406\u8bba\uff08contraction analysis\uff09<\/strong> \u5bf9\u6a21\u578b\u8fdb\u884c\u4e86\u7a33\u5b9a\u6027\u5206\u6790\uff0c\u5e76\u8bc1\u660e\u5728\u4e00\u5b9a\u6761\u4ef6\u4e0b\u7cfb\u7edf\u6ee1\u8db3 <strong>\u03b4ISS\uff08incremental input-to-state stability\uff09<\/strong>\uff0c\u4ece\u800c\u4fdd\u8bc1\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u5728\u52a8\u6001\u73af\u5883\u4e2d\u7684\u7a33\u5b9a\u6027\u3002\u4e3a\u4e86\u8fdb\u4e00\u6b65\u964d\u4f4e\u8ba1\u7b97\u6210\u672c\uff0c\u4f5c\u8005\u8fd8\u63d0\u51fa\u4e86\u4e00\u4e2a <strong>closed-form \u8fd1\u4f3c\u6a21\u578b<\/strong>\uff0c\u901a\u8fc7\u89e3\u6790\u5f62\u5f0f\u8ba1\u7b97\u7cfb\u7edf\u72b6\u6001\uff0c\u4ece\u800c\u907f\u514d\u5728\u6bcf\u4e00\u6b21\u8fed\u4ee3\u4e2d\u6c42\u89e3 ODE\u3002\u8fd9\u79cd\u65b9\u6cd5\u4e0d\u4ec5\u63d0\u5347\u4e86\u8ba1\u7b97\u6548\u7387\uff0c\u540c\u65f6\u51cf\u5c11\u4e86\u8282\u70b9\u4e4b\u95f4\u7684\u901a\u4fe1\u6b21\u6570\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2463\u5b9e\u9a8c\u7ed3\u679c<\/h2>\n\n\n\n<p>\u8bba\u6587\u5728\u4e00\u4e2a\u5178\u578b\u7684 <strong>flocking \u63a7\u5236\u4efb\u52a1<\/strong>\u4e2d\u5bf9\u6a21\u578b\u8fdb\u884c\u4e86\u5b9e\u9a8c\u9a8c\u8bc1\u3002\u5728\u8be5\u4efb\u52a1\u4e2d\uff0c\u4e00\u7ec4\u673a\u5668\u4eba\u9700\u8981\u901a\u8fc7\u5c40\u90e8\u901a\u4fe1\u5b9e\u73b0\u534f\u540c\u8fd0\u52a8\uff1a\u65e2\u8981\u4fdd\u6301\u901f\u5ea6\u4e00\u81f4\uff0c\u53c8\u8981\u907f\u514d\u5f7c\u6b64\u78b0\u649e\uff0c\u540c\u65f6\u7531\u4e00\u4e2a leader \u5f15\u5bfc\u7fa4\u4f53\u5411\u76ee\u6807\u79fb\u52a8\u3002\u5b9e\u9a8c\u5c06 LGTC \u4e0e\u591a\u79cd\u65b9\u6cd5\u8fdb\u884c\u5bf9\u6bd4\uff0c\u5305\u62ec\uff1a<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>GGNN\uff08Gated Graph Neural Network\uff09<\/li>\n\n\n\n<li>GraphODE<\/li>\n\n\n\n<li>CfGC\uff08\u95ed\u5f0f\u8fd1\u4f3c\u7248\u672c\uff09<\/li>\n<\/ul>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"225\" src=\"https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14-1024x225.png\"  class=\"wp-image-1120\" srcset=\"https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14-1024x225.png 1024w, https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14-300x66.png 300w, https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14-768x169.png 768w, https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14-1536x337.png 1536w, https:\/\/www.ndnlab.com\/wp-content\/uploads\/2026\/03\/image-14.png 1544w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" title=\"Liquid-Graph Time-Constant Networkfor Multi-Agent Systems Control\u63d2\u56fe1\" alt=\"Liquid-Graph Time-Constant Networkfor Multi-Agent Systems Control\u63d2\u56fe1\" \/><\/figure>\n\n\n\n<p>\u7ed3\u679c\u8868\u660e\uff0c\u5728\u4e0d\u540c\u89c4\u6a21\u7684\u667a\u80fd\u4f53\u7cfb\u7edf\u4e2d\uff0cLGTC \u5728 <strong>\u7fa4\u4f53\u7f16\u961f\u8bef\u5dee\u548c leader \u76ee\u6807\u8bef\u5dee\u65b9\u9762\u90fd\u4f18\u4e8e\u4f20\u7edf GGNN<\/strong>\u3002\u5177\u4f53\u800c\u8a00\uff0c\u5728\u67d0\u4e9b\u5b9e\u9a8c\u8bbe\u7f6e\u4e0b\uff0cLGTC \u4e0e\u5176\u95ed\u5f0f\u8fd1\u4f3c\u6a21\u578b\u76f8\u6bd4 GGNN <strong>\u5728 flocking \u8bef\u5dee\u4e0a\u53ef\u63d0\u5347\u7ea6 40%\uff0c\u5728 leader \u8bef\u5dee\u4e0a\u63d0\u5347\u7ea6 10%<\/strong>\u3002\u6b64\u5916\uff0c\u5b9e\u9a8c\u8fd8\u9a8c\u8bc1\u4e86\u6a21\u578b\u5728\u4e0d\u540c\u901a\u4fe1\u534a\u5f84\u6761\u4ef6\u4e0b\u7684\u9c81\u68d2\u6027\u3002\u7ed3\u679c\u663e\u793a\uff0c\u5f53\u901a\u4fe1\u8303\u56f4\u53d8\u5316\u65f6\uff0cLGTC \u548c\u5176\u95ed\u5f0f\u8fd1\u4f3c\u6a21\u578b\u4f9d\u7136\u80fd\u591f\u4fdd\u6301\u7a33\u5b9a\u6027\u80fd\uff0c\u5e76\u4e14\u6574\u4f53\u8868\u73b0\u66f4\u63a5\u8fd1\u4e13\u5bb6\u63a7\u5236\u5668\uff08expert controller\uff09\u3002<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2464\u603b\u7ed3<\/h2>\n\n\n\n<p>\u603b\u4f53\u6765\u770b\uff0c\u8fd9\u7bc7\u8bba\u6587\u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684 <strong>\u8fde\u7eed\u65f6\u95f4\u56fe\u795e\u7ecf\u7f51\u7edc\u6a21\u578b LGTC<\/strong>\uff0c\u7528\u4e8e\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u7684\u5206\u5e03\u5f0f\u63a7\u5236\u3002\u8be5\u6a21\u578b\u901a\u8fc7\u7ed3\u5408 <strong>Liquid Time-Constant \u7f51\u7edc\u4e0e\u56fe\u7ed3\u6784\u4fe1\u606f\u4f20\u64ad\u673a\u5236<\/strong>\uff0c\u80fd\u591f\u66f4\u597d\u5730\u523b\u753b\u7f51\u7edc\u5316\u52a8\u6001\u7cfb\u7edf\u7684\u884c\u4e3a\u3002\u540c\u65f6\uff0c\u901a\u8fc7 <strong>\u6536\u7f29\u5206\u6790\u7406\u8bba<\/strong>\u8bc1\u660e\u7cfb\u7edf\u7a33\u5b9a\u6027\uff0c\u5e76\u63d0\u51fa <strong>closed-form \u8fd1\u4f3c\u6a21\u578b<\/strong>\u6765\u964d\u4f4e\u8ba1\u7b97\u4e0e\u901a\u4fe1\u5f00\u9500\u3002\u5b9e\u9a8c\u7ed3\u679c\u8868\u660e\uff0c\u5728\u5178\u578b\u7684\u7fa4\u4f53\u7f16\u961f\u63a7\u5236\u4efb\u52a1\u4e2d\uff0cLGTC \u5728\u9884\u6d4b\u80fd\u529b\u548c\u63a7\u5236\u6027\u80fd\u65b9\u9762\u5747\u4f18\u4e8e\u4f20\u7edf\u56fe\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\u3002\u4ece\u66f4\u5b8f\u89c2\u7684\u89d2\u5ea6\u770b\uff0c\u8fd9\u9879\u5de5\u4f5c\u4f53\u73b0\u4e86\u4e00\u79cd\u503c\u5f97\u5173\u6ce8\u7684\u7814\u7a76\u8d8b\u52bf\uff0c\u5373 <strong>\u8fde\u7eed\u65f6\u95f4\u795e\u7ecf\u7f51\u7edc\u3001\u56fe\u795e\u7ecf\u7f51\u7edc\u4ee5\u53ca\u63a7\u5236\u7406\u8bba\u4e4b\u95f4\u7684\u878d\u5408<\/strong>\u3002\u5bf9\u4e8e\u591a\u673a\u5668\u4eba\u7cfb\u7edf\u3001\u5206\u5e03\u5f0f\u63a7\u5236\u7cfb\u7edf\u4ee5\u53ca\u590d\u6742\u7f51\u7edc\u52a8\u6001\u5efa\u6a21\u800c\u8a00\uff0c\u8fd9\u79cd\u65b9\u6cd5\u53ef\u80fd\u5177\u6709\u5e7f\u6cdb\u7684\u5e94\u7528\u6f5c\u529b\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u2460\u7814\u7a76\u80cc\u666f\uff1a \u5728\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\uff08Multi-Agent Systems, MAS\uff09\u4e2d\uff0c\u901a\u4fe1\u662f\u5b9e\u73b0\u5206\u5e03\u5f0f\u63a7\u5236\u548c\u89c4\u5212\u7684\u5173\u952e\u8981\u7d20\u3002\u901a\u8fc7\u667a\u80fd\u4f53\u4e4b\u95f4\u7684\u4fe1\u606f\u4ea4\u6362\uff0c\u53ef\u4ee5\u5c06\u9884\u6d4b\u548c\u51b3\u7b56\u8fc7\u7a0b\u5206\u5e03\u5230\u591a\u4e2a\u8282\u70b9\u4e0a\uff0c\u4ece\u800c\u63d0\u9ad8\u7cfb\u7edf\u7684\u53ef\u6269\u5c55\u6027\u4e0e\u9884\u6d4b\u80fd\u529b\u3002\u8fd1\u5e74\u6765\uff0c\u968f\u7740\u6570\u636e\u9a71\u52a8\u63a7\u5236\u65b9\u6cd5\u7684\u53d1\u5c55\uff0c\u8d8a\u6765\u8d8a\u591a\u7814\u7a76\u5c1d\u8bd5\u5229\u7528 \u56fe\u795e\u7ecf\u7f51\u7edc\uff08Graph Neural Networks, GNNs\uff09 \u6765\u5efa\u6a21\u591a\u667a\u80fd\u4f53\u7cfb\u7edf\u4e2d\u7684\u4fe1\u606f\u4f20\u64ad\u4e0e\u51b3\u7b56\u8fc7\u7a0b\u3002\u4f8b\u5982\u5df2\u6709\u5de5\u4f5c\u5c06 GNN \u5e94\u7528\u4e8e \u7fa4\u4f53\u7f16\u961f\u63a7\u5236\uff08flocking\uff09\u3001\u7a7a\u95f4\u8986\u76d6\u3001\u591a\u673a\u5668\u4eba\u8def\u5f84\u89c4 &hellip; <a href=\"https:\/\/www.ndnlab.com\/?p=1116\">\u7ee7\u7eed\u9605\u8bfb <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":3,"featured_media":1122,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[],"class_list":["post-1116","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-rengongzhineng"],"_links":{"self":[{"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/posts\/1116","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1116"}],"version-history":[{"count":3,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/posts\/1116\/revisions"}],"predecessor-version":[{"id":1121,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/posts\/1116\/revisions\/1121"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=\/wp\/v2\/media\/1122"}],"wp:attachment":[{"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1116"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1116"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ndnlab.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1116"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}