{"id":34061,"date":"2022-04-25T19:38:32","date_gmt":"2022-04-25T11:38:32","guid":{"rendered":"https:\/\/swarma.org\/?p=34061"},"modified":"2022-04-25T19:38:32","modified_gmt":"2022-04-25T11:38:32","slug":"%e7%bb%bc%e8%bf%b0%e9%80%9f%e9%80%92%ef%bc%9a%e7%bb%88%e8%ba%ab%e5%ad%a6%e4%b9%a0%e7%9a%84%e7%94%9f%e7%89%a9%e5%ad%a6%e5%9f%ba%e7%a1%80","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=34061","title":{"rendered":"\u7efc\u8ff0\u901f\u9012\uff1a\u7ec8\u8eab\u5b66\u4e60\u7684\u751f\u7269\u5b66\u57fa\u7840"},"content":{"rendered":"
\n
\n
\n
\n
\n

<\/p>\n<\/section>\n

\n
\n
\n
\n
\n
\n

<\/section>\n<\/section>\n
\n
\n

\u6458\u8981<\/strong><\/p>\n<\/section>\n<\/section>\n

\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
\n
\n

\u751f\u7269\u6709\u673a\u4f53\u4e00\u751f\u4e2d\u90fd\u4f1a\u4ece\u4e0e\u73af\u5883\u7684\u76f8\u4e92\u4f5c\u7528\u4e2d\u5b66\u4e60\u3002\u4e3a\u4e86\u4f7f\u4eba\u5de5\u667a\u80fd\u7cfb\u7edf\u80fd\u591f\u6210\u529f\u5730\u5728\u73b0\u5b9e\u4e16\u754c\u4e2d\u8fd0\u4f5c\u5e76\u9002\u5e94\uff0c\u4e5f\u9700\u8981\u80fd\u591f\u6301\u7eed\u4e0d\u65ad\u5730\u5b66\u4e60\u3002\u8fd9\u4e00\u6311\u6218\u88ab\u79f0\u4e3a\u7ec8\u8eab\u5b66\u4e60\uff08lifelong learning\uff09\uff0c\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u4ecd\u672a\u89e3\u51b3\u3002\u6700\u8fd1\u53d1\u8868\u4e8e Nature Machine Intelligence<\/em> \u7efc\u8ff0\u6587\u7ae0\u786e\u5b9a\u4e86\u4e00\u7cfb\u5217\u5177\u6709\u7ec8\u8eab\u5b66\u4e60\u80fd\u529b\u7684\u4eba\u5de5\u667a\u80fd\u7cfb\u7edf\u9700\u8981\u5b9e\u73b0\u7684\u5173\u952e\u529f\u80fd\u3002\u4f5c\u8005\u4eec\u63cf\u8ff0\u4e86\u4e00\u4e9b\u751f\u7269\u5b66\u673a\u5236\uff0c\u5305\u62ec\u795e\u7ecf\u5143\u548c\u975e\u795e\u7ecf\u5143\u5c42\u9762\u7684\uff0c\u8fd9\u4e9b\u673a\u5236\u6709\u52a9\u4e8e\u89e3\u91ca\u751f\u7269\u4f53\u5982\u4f55\u5e94\u5bf9\u7ec8\u8eab\u5b66\u4e60\u7684\u6311\u6218\u3002\u6587\u7ae0\u901a\u8fc7\u793a\u4f8b\u5c55\u793a\u751f\u7269\u542f\u53d1\u6a21\u578b\uff0c\u5e76\u9610\u8ff0\u4e86\u7528\u4e8e\u4eba\u5de5\u667a\u80fd\u4ee5\u4fc3\u8fdb\u5177\u6709\u7ec8\u8eab\u5b66\u4e60\u673a\u5668\u5f00\u53d1\u7684\u751f\u7269\u5b66\u4e0a\u53ef\u884c\u7684\u673a\u5236\u3002\u6700\u540e\u8ba8\u8bba\u4e86\u8fdb\u4e00\u6b65\u4e86\u89e3\u548c\u4fc3\u8fdb\u7ec8\u751f\u5b66\u4e60\u6280\u672f\u6c34\u5e73\u7684\u673a\u4f1a\uff0c\u76ee\u7684\u662f\u51cf\u5c0f\u81ea\u7136\u4e0e\u4eba\u5de5\u667a\u80fd\u95f4\u7684\u5dee\u8ddd\u3002<\/span><\/strong><\/span><\/p>\n

 <\/span><\/p>\n

\u7814\u7a76\u9886<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong>\u57df\uff1a<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong>\u7c7b\u8111\u8ba1\u7b97\uff0c\u7ec8\u8eab\u5b66\u4e60\uff0c\u8ba1\u7b97\u795e\u7ecf\u5b66\uff0c\u707e\u96be\u6027\u9057\u5fd8<\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/span><\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

\n
\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n
\n
<\/strong><\/section>\n<\/section>\n
\n

\u90ed\u745e\u4e1c<\/span> <\/span>| \u4f5c\u8005<\/strong><\/span><\/p>\n

\u9093\u4e00\u96ea<\/span> | \u7f16\u8f91<\/strong><\/span><\/strong><\/span><\/p>\n


<\/p>\n<\/section>\n


<\/p>\n

<\/section>\n
\n
\n
\n

\u8bba\u6587\u9898\u76ee\uff1a<\/span><\/p>\n

Biological underpinnings for lifelong learning machines<\/span><\/p>\n

\n
\u8bba\u6587\u94fe\u63a5\uff1a<\/span><\/section>\n
https:\/\/www.nature.com\/articles\/s42256-022-00452-0<\/span><\/section>\n<\/h3>\n<\/section>\n<\/section>\n<\/blockquote>\n

\n

<\/section>\n
\u7ec8\u8eab\u5b66\u4e60<\/span>\uff08Life-long learning\uff09<\/span>\u6307\u667a\u80fd\u4f53\u80fd\u591f\u6301\u7eed\u4ece\u4e0e\u73af\u5883\u7684\u4e92\u52a8\u63d0\u53d6\u4fe1\u606f\u7684\u80fd\u529b\uff0c\u5bf9\u4e8e\u751f\u7269\u4f53\u6765\u8bf4\uff0c\u8fd9\u662f\u4e60\u4ee5\u4e3a\u5e38\u7684\u4e8b\uff0c\u4f46\u5bf9\u4e8e\u4eba\u5de5\u667a\u80fd\u6765\u8bf4\uff0c\u5374\u5f88\u56f0\u96be\u3002\u5177\u6709\u7ec8\u8eab\u5b66\u4e60\u80fd\u529b\u7684\u7cfb\u7edf\u5177\u6709\u5f7b\u5e95\u6539\u53d8\u8bb8\u591a\u5e94\u7528\u7a0b\u5e8f\u7684\u6f5c\u529b\uff0c\u5305\u62ec\u5168\u81ea\u52a8\u9a7e\u9a76\u6c7d\u8f66\u3001\u667a\u80fd\u57ce\u5e02\u548c\u533b\u7597\u4fdd\u5065\u3002\u53d1\u6398\u8fd9\u4e00\u6f5c\u529b\u5c06\u9700\u8981\u5f00\u5c55\u591a\u5b66\u79d1\u5408\u4f5c\uff0c\u4ee5\u652f\u6301\u751f\u7269\u5b66\u3001\u795e\u7ecf\u79d1\u5b66\u3001\u5fc3\u7406\u5b66\u3001\u5de5\u7a0b\u548c\u4eba\u5de5\u667a\u80fd\u4ea4\u53c9\u9886\u57df\u7684\u7814\u7a76\u4eba\u5458\u5408\u4f5c\u4ea7\u751f\u53ef\u843d\u5730\u7684\u6210\u679c\u3002Nature Machine Intelligence<\/em> \u8fd1\u671f\u7684\u7efc\u8ff0\u8bba\u6587\uff0c\u6307\u51fa\u4e86\u7ec8\u751f\u5b66\u4e60\u6240\u9700\u8981\u8fbe\u6210\u7684\u5177\u4f53\u76ee\u6807\uff0c\u4ee5\u53ca\u4eba\u5de5\u667a\u80fd\u7684\u7814\u7a76\u8005\u80fd\u4ece\u751f\u7269\u5b66\u7814\u7a76\u4e2d\u83b7\u5f97\u54ea\u4e9b\u501f\u9274\u548c\u542f\u53d1\u3002<\/span><\/section>\n
 <\/span><\/section>\n<\/h3>\n

\n
 <\/span>
<\/section>\n<\/h1>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

1. \u7ec8\u8eab\u5b66\u4e60\u7684\u516d\u4e2a\u5173\u952e\u7279\u5f81<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n

\u5f53\u6211\u4eec\u8bf4\u67d0\u9879\u7b97\u6cd5\u6539\u8fdb\u63d0\u5347\u4e86\u6a21\u578b\u7ec8\u8eab\u5b66\u4e60\u7684\u80fd\u529b\u65f6\uff0c\u53ef\u4ee5\u5c06\u5176\u5f52\u4e3a\u4ee5\u4e0b6\u7c7b\u4e2d\u67d0\u4e00\u7c7b\u6216\u5176\u7ec4\u5408\uff0c\u5206\u522b\u662f\uff1a\u4efb\u52a1\u8fc1\u79fb\u548c\u9002\u5e94\u3001\u514b\u670d\u707e\u96be\u6027\u9057\u5fd8\u3001\u5229\u7528\u4efb\u52a1\u76f8\u4f3c\u6027\u3001\u5177\u6709\u4efb\u52a1\u65e0\u5173\u7684\u5b66\u4e60\u80fd\u529b\u3001\u5bf9\u566a\u97f3\u7684\u5bb9\u5fcd\u3001\u4ee5\u53ca\u8d44\u6e90\u9ad8\u6548\u5229\u7528\u548c\u53ef\u6301\u7eed\u6027\u3002<\/span><\/section>\n
                 <\/span><\/section>\n
<\/section>\n

\u56fe1. \u7ec8\u8eab\u5b66\u4e60\u6240\u9700\u76846\u7c7b\u5173\u952e\u7279\u5f81<\/span><\/p>\n


<\/span><\/strong><\/p>\n

\u4efb\u52a1\u8fc1\u79fb<\/span><\/strong>\uff1a\u5c06\u4ee5\u524d\u7684\u77e5\u8bc6\u5e94\u7528\u4e8e\u65b0\u4efb\u52a1\u5e76\u8fc5\u901f\u9002\u5e94\u4efb\u52a1\u6216\u73af\u5883\u53d8\u5316\u7684\u80fd\u529b\u3002\u4e0a\u56fe\u4e2d\uff0c\u667a\u80fd\u4f53\u5728\u4efb\u52a1 B \uff08\u5305\u88c5\u7bb1\u4e2d\u7684\u7269\u4f53\uff09\u8fdb\u884c\u8bad\u7ec3\uff0c\u5e76\u968f\u540e\u80fd\u591f\u5e94\u7528\u5b66\u5230\u7684\u6280\u80fd\uff0c\u4ee5\u4fc3\u8fdb\u5b66\u4e60\u7c7b\u4f3c\u4f46\u4e0d\u540c\u53d8\u4f53\u7684\u4efb\u52a1\uff08\u5bf9\u8c61\u548c\u76d2\u5b50\u7684\u4e0d\u540c\u5927\u5c0f\u548c\u5f62\u72b6\uff09\u3002<\/span><\/p>\n


<\/span><\/p>\n

\u514b\u670d\u707e\u96be\u6027\u9057\u5fd8<\/span><\/strong>\uff1a\u4e0d\u4f1a\u56e0\u4e3a\u5b66\u4e60\u65b0\u7684\u4efb\u52a1\u800c\u964d\u4f4e\u4e4b\u524d\u5b66\u4e60\u7684\u4efb\u52a1\u4e0a\u7684\u8868\u73b0\u3002<\/span><\/p>\n

<\/span><\/p>\n

\u5229\u7528\u4efb\u52a1\u76f8\u4f3c\u6027<\/span><\/strong>\uff1a\u4e0d\u9700\u8981\u5b66\u4e60\u4efb\u52a1\u7684\u5355\u4e00\u8868\u793a\uff0c\u800c\u662f\u80fd\u591f\u5c06\u4efb\u52a1\u5206\u89e3\u4e3a\u5b66\u4e60\u65b0\u4efb\u52a1\u65f6\u53ef\u4ee5\u5e94\u7528\u7684\u5b50\u4efb\u52a1\u3002\u4e0a\u56fe\u4e2d\u4f5c\u4e3a\u4efb\u52a1B\u8bad\u7ec3\u7684\u4e00\u90e8\u5206\u6240\u5b66\u4e60\u7684\u5b9a\u4f4d\u52a8\u4f5c\u53ef\u4ee5\u76f4\u63a5\u8f6c\u79fb\u5230\u4efb\u52a1C\u4e0a\uff0c\u53ef\u91cd\u590d\u4f7f\u7528\u8be5\u6280\u80fd\u3002<\/span><\/p>\n


<\/span><\/p>\n

\u4efb\u52a1\u65e0\u5173\u7684\u5b66\u4e60<\/span><\/strong>\uff1a\u80fd\u89e3\u51b3\u4e00\u4e2a\u95ee\u9898\uff0c\u800c\u4e0d\u5fc5\u660e\u786e\u5730\u544a\u8bc9\u667a\u80fd\u4f53\u8fd9\u4e2a\u95ee\u9898\u5c5e\u4e8e\u51e0\u4e2a\u5b66\u4e60\u4efb\u52a1\u4e2d\u7684\u54ea\u4e00\u4e2a\u3002<\/span><\/p>\n


<\/span><\/p>\n

\u5bf9\u566a\u97f3\u7684\u5bb9\u5fcd<\/span><\/strong>\uff1a\u667a\u80fd\u7cfb\u7edf\u5728\u6ca1\u6709\u4efb\u4f55\u5e72\u6270\u7684\u60c5\u51b5\u4e0b\u88ab\u8bad\u7ec3\u6267\u884c\u4efb\u52a1\uff0c\u968f\u540e\u80fd\u591f\u5728\u73b0\u5b9e\u4e16\u754c\u5ffd\u7565\u4e0d\u76f8\u5173\u7684\u4e8b\u7269\u548c\u6f5c\u5728\u7684\u5206\u6563\u6ce8\u610f\u529b\u7684\u884c\u4e3a\u3002<\/span><\/p>\n


<\/span><\/p>\n

\u8d44\u6e90\u9ad8\u6548\u5229\u7528\u548c\u53ef\u6301\u7eed\u6027<\/span><\/strong>\uff1a\u5229\u7528\u6709\u9650\u7684\u8d44\u6e90\uff0c\u4e0d\u65ad\u5b66\u4e60\u65b0\u77e5\u8bc6\u7684\u80fd\u529b\u3002<\/span><\/section>\n
 <\/span><\/section>\n

 <\/span><\/p>\n

\n
\n
\n
\n
\n

<\/section>\n

<\/section>\n
\n
<\/svg><\/section>\n<\/section>\n<\/section>\n
\n

2. \u542f\u53d1\u7ec8\u8eab\u5b66\u4e60\u7684\u4e5d\u79cd\u751f\u7269\u5b66\u673a\u7406<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


<\/p>\n

\u5728\u8fd9\u7bc7\u6709\u4f17\u591a\u4f5c\u8005\u64b0\u5199\u7684\u7efc\u8ff0\u4e2d\uff0c\u66f4\u591a\u7684\u7bc7\u5e45\u82b1\u5728\u4ecb\u7ecd\u795e\u7ecf\u79d1\u5b66\u7684\u7814\u7a76\u6210\u679c\uff0c\u8fd8\u6709\u8fd9\u4e9b\u6210\u679c\u76ee\u524d\u548c\u4eba\u5de5\u667a\u80fd\u9886\u57df\u5df2\u7ecf\u6216\u8005\u53ef\u80fd\u64e6\u51fa\u7684\u706b\u82b1<\/span>\uff08\u4e0d\u540c\u751f\u7269\u5b66\u673a\u7406\u7684\u5e94\u7528\u5728\u4e0b\u56fe\u7684\u6587\u732e\u4e2d\u5177\u4f53\u9610\u8ff0\uff09<\/span>\u3002<\/span><\/section>\n
 <\/span><\/section>\n
\u4e0b\u56fe\u6bcf\u4e00\u884c\u662f\u4e00\u79cd\u5927\u8111\u8fd0\u884c\u673a\u5236\uff0c\u6709\u6570\u5b57\u7684\u4f4d\u7f6e\u8868\u660e\u53ef\u4ee5\u5728\u8be5\u4efb\u52a1\u4e0a\u4f7f\u7528\u7c7b\u4f3c\u7684\u751f\u7269\u5b66\u539f\u7406\u542f\u53d1\u7ec8\u8eab\u5b66\u4e60\u7684\u8bbe\u8ba1\u3002\u6bcf\u4e2a\u6570\u5b57\u5bf9\u5e94\u4e00\u4e2a\u6587\u732e\u3002\u53f3\u56fe\u662f\u5728\u8ba1\u7b97\u673a\u89c6\u89c9\uff0c\u73af\u5883\u4ea4\u4e92\u3001\u751f\u7269\u4f53\u6a21\u62df\u3001\u673a\u5668\u4eba\u9886\u57df\u76f8\u5173\u7684\u8bc4\u4ef7\u6570\u636e\u96c6\u6307\u6807\u5bf9\u5e94\u7684\u6587\u732e\u3002<\/span><\/section>\n

<\/span><\/section>\n
<\/section>\n

\u56fe2. \u751f\u7269\u542f\u53d1\u6a21\u578b\u7684\u5e94\u7528\u3002<\/span><\/p>\n

    \n
  1. \n

    \u795e\u7ecf\u5f62\u6210\uff08neurogenesis\uff09\uff0c\u6307\u5927\u8111\u4e2d\u53ef\u4ee5\u65b0\u4ea7\u751f\u795e\u7ecf\u5143\u7684\u80fd\u529b\u3002\u5728\u523a\u6fc0\u4e30\u5bcc\u7684\u73af\u5883\u4e2d\u957f\u5927\u7684\u52a8\u7269\uff0c\u5176\u5927\u8111\u4e2d\u7684\u795e\u7ecf\u5143\u7684\u751f\u6210\u901f\u5ea6\u4e5f\u66f4\u5feb\u3002<\/span><\/p>\n<\/li>\n

  2. \n

    \u4e8b\u4ef6\u8bb0\u5fc6\u7684\u56de\u653e\uff08episodic replay\uff09\uff0c\u6307\u4eba\u4f1a\u5728\u505a\u68a6\u65f6\u56de\u653e\u767d\u5929\u7ecf\u5386\u7684\u4e8b\uff0c\u8fd9\u6837\u7684\u56de\u653e\u6700\u521d\u51fa\u73b0\u5728\u6d77\u9a6c\u4f53\uff0c\u4e4b\u540e\u5728\u6d77\u9a6c\u4f53\u548c\u65b0\u76ae\u5c42\u533a\u57df\u540c\u6b65\u51fa\u73b0\u3002<\/span><\/p>\n<\/li>\n

  3. \n

    \u5143\u53ef\u5851\u6027\uff08metaplasticity\uff09\u6307\u795e\u7ecf\u5143\u95f4\u7684\u7a81\u89e6\u5f3a\u5ea6\u5177\u6709\u53cc\u91cd\u53ef\u5851\u6027\uff0c\u4e0d\u4ec5\u53d7\u5230\u5468\u56f4\u795e\u7ecf\u5143\u7684\u5f71\u54cd\uff0c\u8fd8\u53d7\u5230\u81ea\u8eab\u795e\u7ecf\u5143\u5185\u90e8\u751f\u5316\u6d3b\u52a8\u7684\u5f71\u54cd\u3002<\/span><\/p>\n<\/li>\n

  4. \n

    \u795e\u7ecf\u8c03\u5236\uff08Neuromodulation\uff09\u6307\u8c03\u8282\u6027\u795e\u7ecf\u5143\u91ca\u653e\u7684\u795e\u7ecf\u9012\u8d28\uff08\u5982\u591a\u5df4\u80fa\u3001\u50ac\u4ea7\u7d20\uff09\uff0c\u4e0d\u4ec5\u80fd\u591f\u5f71\u54cd\u5c40\u90e8\u7684\u795e\u7ecf\u6d3b\u52a8\u548c\u53ef\u5851\u6027\uff0c\u8fd8\u80fd\u4ea7\u751f\u5168\u5c40\u6027\u7684\u5f71\u54cd\u3002<\/span><\/p>\n<\/li>\n

  5. \n

    \u80cc\u666f\u4f9d\u8d56\u7684\u611f\u77e5\uff08Context-dependent perception and gating\uff09\uff1a\u751f\u7269\u4f53\u80fd\u591f\u6839\u636e\u80cc\u666f\u4fe1\u606f\u9009\u62e9\u6027\u5730\u6295\u5c04\u6ce8\u610f\u529b\uff0c\u5e76\u6574\u5408\u65b0\u4fe1\u606f\uff0c\u8fd8\u53ef\u4ee5\u6fc0\u6d3b\u90e8\u5206\u795e\u7ecf\u5143\u3002\u9762\u5bf9\u81ea\u4e0a\u800c\u4e0b\u7684\u4efb\u52a1\uff0c\u5927\u8111\u80fd\u9ad8\u6548\u5730\u5206\u914d\u6ce8\u610f\u529b\u3002<\/span><\/p>\n<\/li>\n

  6. \n

    \u5c42\u7ea7\u5316\u7684\u5206\u5e03\u5f0f\u5904\u7406\uff1a\u8bb8\u591a\u751f\u7269\u4f53\u6ca1\u6709\u6216\u53ea\u6709\u6781\u5c0f\u7684\u96c6\u4e2d\u7684\u5927\u8111\uff0c\u4f46\u5374\u6709\u5f88\u591a\u5c40\u90e8\u7684\u795e\u7ecf\u5904\u7406\u6a21\u5757\u3002\u901a\u8fc7\u5c42\u7ea7\u5316\u7684\u5206\u5e03\u63a7\u5236\uff0c\u4f7f\u5f97\u4fe1\u53f7\u5904\u7406\u548c\u5b66\u4e60\u53ef\u4ee5\u5206\u5e03\u5728\u5168\u8eab\u7684\u591a\u4e2a\u795e\u7ecf\u5143\u7f51\u7edc\u4e2d\uff0c\u6bcf\u4e2a\u795e\u7ecf\u5143\u90fd\u6709\u8f83\u9ad8\u7684\u7f51\u7edc\u5185\u8fde\u63a5\uff0c\u4ee5\u53ca\u76f8\u5bf9\u7a00\u758f\u7684\u7f51\u7edc\u95f4\u8fde\u63a5\u3002<\/span><\/p>\n<\/li>\n

  7. \n

    \u5927\u8111\u5916\u7684\u8ba4\u77e5\u6d3b\u52a8\uff1a\u4e0d\u5177\u5907\u5927\u8111\u7684\u751f\u7269\uff0c\u8fd8\u5305\u62ec\u5355\u7ec6\u80de\u751f\u7269\u751a\u81f3\u662f\u751f\u7269\u4f53\u4e2d\u7684\u5206\u5b50\u7f51\u7edc\uff0c\u8868\u73b0\u51fa\u4ece\u7ecf\u9a8c\u4e2d\u5b66\u4e60\u3001\u9884\u6d4b\u672a\u6765\u4e8b\u4ef6\u548c\u5bf9\u65b0\u7684\u6311\u6218\u505a\u51fa\u9002\u5e94\u7684\u80fd\u529b\u3002\u8bf8\u5982\u751f\u7269\u7535\u3001\u591a\u5c3a\u5ea6\u7684\u7a33\u6001\u673a\u5236\u3001\u7ec4\u7ec7\u5185\u90e8\u548c\u8de8\u5c42\u6b21\u7684\u5408\u4f5c\u548c\u7ade\u4e89\u662f\u5176\u751f\u7269\u5b66\u57fa\u7840\u3002<\/span><\/p>\n<\/li>\n

  8. \n

    \u53ef\u91cd\u5851\u7684\u6709\u673a\u4f53\uff1a\u751f\u7269\u4f53\u662f\u9ad8\u5ea6\u53ef\u91cd\u6784\u7684\uff0c\u5c3d\u7ba1\u73af\u5883\u548c\u7ec6\u80de\u7279\u6027\u6025\u5267\u53d8\u5316\uff0c\u5b83\u4eec\u4ecd\u7136\u4fdd\u6301\u8fde\u8d2f\u3001\u9002\u5e94\u6027\u7684\u529f\u80fd\u3002\u800c\u5177\u6709\u8fd9\u4e00\u80fd\u529b\u7684\u7b97\u6cd5\uff0c\u5c06\u80fd\u5229\u7528\u5df2\u6709\u4fe1\u606f\u9002\u5e94\u65b0\u7684\u573a\u666f\u3002<\/span><\/p>\n<\/li>\n

  9. \n

    \u591a\u611f\u5b98\u8f93\u5165\u6574\u5408\uff1a\u751f\u7269\u4f53\u672c\u8d28\u4e0a\u662f\u4e00\u79cd\u611f\u89c9\u8fd0\u52a8\u7cfb\u7edf\uff0c\u5176\u884c\u4e3a\u7531\u591a\u79cd\u7c7b\u578b\u7684\u611f\u89c9\u4fe1\u53f7\u51b3\u5b9a\u3002\u7406\u89e3\u751f\u7269\u4f53\u5982\u4f55\u4ee5\u4e00\u79cd\u901a\u7528\u7684\u4e0e\u4efb\u52a1\u65e0\u5173\u7684\u65b9\u5f0f\u5904\u7406\u3001\u8fc7\u6ee4\u548c\u5904\u7406\u5927\u91cf\u7684\u611f\u5b98\u6570\u636e\uff0c\u53ef\u4ee5\u652f\u6301\u7ec8\u8eab\u5b66\u4e60\u7b97\u6cd5\u7684\u5f00\u53d1\u3002<\/span><\/p>\n<\/li>\n<\/ol>\n

     <\/span><\/section>\n
    \u9664\u6b64\u4e4b\u5916\uff0c\u52a8\u6001\u8bb0\u5fc6\u66f4\u65b0\u673a\u5236\uff0c\u5982\u4e3b\u52a8\u9057\u5fd8<\/span>\uff08active forgeting\uff09<\/span>\u3001\u8bb0\u5fc6\u64e6\u9664<\/span>\uff08extinction\uff09<\/span>\u548c\u8bb0\u5fc6\u518d\u5de9\u56fa<\/span>\uff08memory reconsolidation\uff09<\/span>\u3001\u57fa\u56e0\u95f4\u7684\u4fe1\u53f7\u901a\u8def\u548c\u8c03\u63a7\u3001\u7ec6\u80de\u95f4\u901a\u8baf\u7b49\u53d1\u73b0\uff0c\u4e5f\u53ef\u4ee5\u6210\u4e3a\u7ec8\u8eab\u5b66\u4e60\u7684\u7075\u611f\u6765\u6e90\u3002<\/span><\/section>\n

    <\/section>\n

     <\/span><\/p>\n

    \n
    \n
    \n
    \n
    \n

    <\/section>\n

    <\/section>\n
    \n
    <\/svg><\/section>\n<\/section>\n<\/section>\n
    \n

    3. \u603b\u7ed3<\/span><\/strong><\/strong><\/p>\n<\/section>\n

    \n

    <\/section>\n

    <\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n


    <\/p>\n

    \u53d7\u751f\u7269\u542f\u53d1\u7684\u6a21\u578b\u5e94\u7528\u4e8e\u7ec8\u8eab\u5b66\u4e60\uff0c\u8be5\u7efc\u8ff0\u63d0\u4f9b\u4e86\u4e00\u4e9b\u8be6\u5c3d\u7684\u6848\u4f8b\uff0c\u8bf4\u660e\u8fd9\u4e9b\u65b9\u6cd5\u53ef\u4ee5\u8d85\u8d8a\u5f53\u524d\u4eba\u5de5\u667a\u80fd\u7684\u5c40\u9650\u6027\u3002\u5c3d\u7ba1\u8bb8\u591a\u8fd9\u4e9b\u5f00\u53d1\u4ecd\u5904\u4e8e\u8d77\u6b65\u9636\u6bb5\uff0c\u5305\u62ec\u5bf9\u5404\u4e2a\u7279\u6027\u5728\u5c0f\u6570\u636e\u96c6\u4e0a\uff0c\u5173\u4e8e\u5b9e\u73b0\u7279\u5b9a\u7ec8\u8eab\u5b66\u4e60\u76ee\u6807\u7684\u6027\u80fd\u5c55\u793a\u3002\u672a\u6765\u4eba\u5de5\u667a\u80fd\u793e\u533a\u901a\u8fc7\u53d7\u751f\u7269\u542f\u53d1\u7684\u673a\u5236\uff0c\u53ef\u6539\u8fdb\u7b97\u6cd5\uff0c\u5e76\u5c06\u5176\u6574\u5408\u5230\u6807\u51c6\u7684\u673a\u5668\u5b66\u4e60\u6846\u67b6\u4e2d<\/strong>\uff0c\u4ece\u800c\u5f00\u53d1\u5177\u6709\u66f4\u5927\u81ea\u4e3b\u6027\u548c\u7ec8\u8eab\u5b66\u4e60\u80fd\u529b\u7684\u65b0\u4e00\u4ee3\u4eba\u5de5\u667a\u80fd\u7cfb\u7edf\u3002<\/span><\/section>\n
     <\/span><\/section>\n
    \u4e0a\u8ff0\u76ee\u6807\u6210\u529f\u7684\u5173\u952e\uff0c\u5728\u4e8e\u6df7\u5408\u4f7f\u7528\u4e0a\u9762\u5217\u51fa\u7684\u51e0\u79cd\u673a\u5236<\/span>\uff08\u4ee5\u53ca\u5173\u6ce8\u751f\u7269\u5b66\u53ca\u795e\u7ecf\u79d1\u5b66\u7684\u8fdb\u5c55\uff0c\u53ca\u65f6\u6574\u5408\u90a3\u4e9b\u5c1a\u672a\u53d1\u73b0\u7684\u673a\u5236\uff09<\/span>\uff0c\u800c\u4e0d\u662f\u72ed\u9698\u5730\u5173\u6ce8\u67d0\u4e00\u9879\u673a\u5236\u3002\u53e6\u4e00\u4e2a\u5173\u952e\u56e0\u7d20\u662f\u5f00\u53d1\u73b0\u5b9e\u7684\u6d4b\u8bd5\u73af\u5883\uff0c\u8fd9\u4e9b\u73af\u5883\u4e13\u95e8\u9488\u5bf9\u7ec8\u8eab\u5b66\u4e60\u80fd\u529b\u8bbe\u8ba1\uff0c\u800c\u4e0d\u662f\u8ba9\u7b97\u6cd5\u4e0d\u5c40\u9650\u4e8e\u9884\u5148\u51c6\u5907\u7684\u6570\u636e\u96c6\u3002\u901a\u8fc7\u521b\u9020\u65b0\u7684\u786c\u4ef6\uff0c\u4f8b\u5982\u795e\u7ecf\u5f62\u6001\u82af\u7247\uff0c\u901a\u8fc7\u8f6f\u786c\u4ef6\u7ed3\u5408\uff0c\u5b9e\u73b0\u795e\u7ecf\u50a8\u5907\u6c60\u8ba1\u7b97<\/span>\uff08reservoir computing\uff09<\/span>\uff0c\u6216\u8109\u51b2\u795e\u7ecf\u7f51\u7edc<\/span>\uff08spike neuron network\uff09<\/span>\u7684\u52a0\u901f\uff0c\u4e5f\u662f\u5fc5\u9700\u7684\u6b65\u9aa4\u3002<\/span><\/section>\n

    <\/section>\n

    <\/section>\n
    \n
    \n
    \n

    \u8bba\u6587 Abstract<\/strong><\/span><\/p>\n<\/section>\n<\/section>\n<\/section>\n

    Biological organisms learn from interactions with their environment throughout their lifetime. For artificial systems to successfully act and adapt in the real world, it is desirable to similarly be able to learn on a continual basis. This challenge is known as lifelong learning, and remains to a large extent unsolved. In this Perspective article, we identify a set of key capabilities that artificial systems will need to achieve lifelong learning. We describe a number of biological mechanisms, both neuronal and non-neuronal, that help explain how organisms solve these challenges, and present examples of biologically inspired models and biologically plausible mechanisms that have been applied to artificial systems in the quest towards development of lifelong learning machines. We discuss opportunities to further our understanding and advance the state of the art in lifelong learning, aiming to bridge the gap between natural and artificial intelligence.<\/span><\/section>\n

    <\/section>\n
     <\/span><\/section>\n
    \n
    \n
    \n
    \n

    \u590d\u6742\u79d1\u5b66\u6700\u65b0\u8bba\u6587<\/strong><\/p>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n

    \n


    <\/p>\n<\/section>\n

    \u96c6\u667a\u6591\u56fe\u9876\u520a\u8bba\u6587\u901f\u9012\u680f\u76ee\u4e0a\u7ebf\u4ee5\u6765\uff0c\u6301\u7eed\u6536\u5f55\u6765\u81eaNature\u3001Science<\/span>\u7b49\u9876\u520a\u7684\u6700\u65b0\u8bba\u6587\uff0c\u8ffd\u8e2a\u590d\u6742\u7cfb\u7edf\u3001\u7f51\u7edc\u79d1\u5b66\u3001\u8ba1\u7b97\u793e\u4f1a\u79d1\u5b66\u7b49\u9886\u57df\u7684\u524d\u6cbf\u8fdb\u5c55\u3002\u73b0\u5728\u6b63\u5f0f\u63a8\u51fa\u8ba2\u9605\u529f\u80fd\uff0c\u6bcf\u5468\u901a\u8fc7\u5fae\u4fe1\u670d\u52a1\u53f7\u300c\u96c6\u667a\u6591\u56fe\u300d\u63a8\u9001\u8bba\u6587\u4fe1\u606f\u3002\u626b\u63cf\u4e0b\u65b9\u4e8c\u7ef4\u7801\u5373\u53ef\u4e00\u952e\u8ba2\u9605\uff1a<\/span>
    <\/p>\n

    <\/p>\n

    \n


    <\/p>\n<\/section>\n


    <\/span><\/p>\n

    \n

    \u63a8\u8350\u9605\u8bfb<\/span><\/strong><\/strong><\/p>\n