{"id":39515,"date":"2022-12-27T20:06:24","date_gmt":"2022-12-27T12:06:24","guid":{"rendered":"https:\/\/swarma.org\/?p=39515"},"modified":"2022-12-27T20:06:24","modified_gmt":"2022-12-27T12:06:24","slug":"%e7%ac%ac%e4%ba%8c%e5%b1%8a%e7%a5%9e%e7%bb%8f%e7%bd%91%e7%bb%9c%e6%96%b0%e5%b9%b4%e4%ba%a4%e5%8f%89%e8%ae%ba%e5%9d%9b","status":"publish","type":"post","link":"https:\/\/swarma.org\/?p=39515","title":{"rendered":"\u7b2c\u4e8c\u5c4a\u795e\u7ecf\u7f51\u7edc\u65b0\u5e74\u4ea4\u53c9\u8bba\u575b"},"content":{"rendered":"
\n
<\/section>\n

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

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

\u8bba\u575b\u7b80\u4ecb<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n
 <\/span><\/section>\n
\u6df1\u5ea6\u795e\u7ecf\u7f51\u7edc\u6b63\u5728\u6210\u4e3aAI for Science\u7684\u57fa\u672c\u5de5\u5177\uff0c\u6784\u6210\u65b0\u7684\u79d1\u5b66\u7814\u7a76\u8303\u5f0f\uff0c\u4ee5\u524d\u6240\u672a\u6709\u7684\u5e7f\u5ea6\u548c\u6df1\u5ea6\u5f71\u54cd\u79d1\u5b66\u7684\u53d1\u5c55\u3002\u7834\u89e3\u6df1\u5ea6\u7f51\u7edc\u7684\u673a\u5236\u4f7f\u5f97\u4eba\u4eec\u80fd\u4ece\u7b2c\u4e00\u6027\u539f\u7406\u51fa\u53d1\u8bbe\u8ba1\u66f4\u597d\u7684\u7f51\u7edc\u7ed3\u6784\u548c\u7b97\u6cd5\uff0c\u751a\u81f3\u6709\u52a9\u4e8e\u7406\u89e3\u751f\u7269\u667a\u80fd\u7684\u672c\u8d28\u3002\u795e\u7ecf\u7f51\u7edc\u65b0\u5e74\u4ea4\u53c9\u8bba\u575b\u4e8e2021\u5e74\u5e95\u7531\u4e00\u7fa4\u70ed\u8877\u4e8e\u4ea4\u53c9\u5b66\u79d1\u7814\u7a76\u7684\u9752\u5e74\u79d1\u5b66\u5bb6\u521b\u7acb\uff0c\u9080\u8bf7\u6765\u81ea\u4e0d\u540c\u5b66\u79d1\u80cc\u666f\u7684\u79d1\u5b66\u5bb6\u8fdb\u884c\u4eba\u5de5\uff08\u751f\u7269\uff09\u667a\u80fd\u539f\u7406\u65b9\u9762\u7684\u5e74\u5ea6\u7814\u8ba8\u3002<\/span><\/section>\n

<\/section>\n
\u7b2c\u4e8c\u5c4a\u795e\u7ecf\u7f51\u7edc\u65b0\u5e74\u4ea4\u53c9\u8bba\u575b\u5c06\u4e8e2022\u5e7412\u670831\uff5e2023\u5e741\u67081\u65e5\u4e3e\u884c\uff08\u7ebf\u4e0a\uff09\u3002\u8be5\u8bba\u575b\u5df2\u6210\u529f\u4e3e\u529e\u4e86\u7b2c\u4e00\u5c4a <\/span>(\u8532\u4eab\u94fe\u63a5\uff1ahttps:\/\/www.koushare.com\/lives\/room\/506610) \u3002\u7b2c\u4e8c\u5c4a\u5c06\u9080\u8bf7\u5728\u4eba\u5de5\u667a\u80fd\uff0c\u6570\u5b66\u7269\u7406\u548c\u8111\u79d1\u5b66\u4ea4\u53c9\u9886\u57df\u4ece\u4e8b\u57fa\u7840\u7814\u7a76\u7684\u79d1\u5b66\u5bb6\u5206\u4eab\u9886\u57df\u524d\u6cbf\u548c\u672a\u6765\u8d8b\u52bf\u5c55\u671b\u3002<\/span><\/section>\n

<\/section>\n

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

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

\u7ec4\u7ec7\u5355\u4f4d<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n
 <\/span><\/section>\n
\u8bba\u575b\u4e3b\u529e\u5355\u4f4d\uff1a\u4e2d\u5c71\u5927\u5b66\u7269\u7406\u5b66\u9662\u3001\u5fc3\u7406\u5b66\u7cfb<\/span><\/section>\n
\u8bba\u575b\u534f\u529e\u5355\u4f4d\uff1a\u66a8\u5357\u5927\u5b66\u7269\u7406\u5b66\u7cfb<\/span><\/section>\n

<\/section>\n
\u8bba\u575b\u7ec4\u59d4\u4f1a\uff1a\u9ec4\u6d77\u5e73\u3001\u5e93\u9038\u8f69\u3001\u5f20\u5e0c\u6600<\/span><\/section>\n

<\/section>\n
\u8bba\u575b\u5b66\u672f\u59d4\u5458\uff1a\u6bd5\u5219\u680b\u3001\u9648\u56fd\u748b\u3001\u9ec4\u6d77\u5e73\u3001\u96f7\u6cfd\u3001\u9ece\u52c3\u3001\u5e93\u9038\u8f69\u3001\u5b5f\u7965\u660e\u3001\u9b4f\u5baa\uff0c\u4f59\u8087\u98de\u3001\u6768\u51ac\u5e73\uff0c\u5f20\u6587\u660a\u3001\u5f20\u5e0c\u6600<\/span><\/section>\n

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

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

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

\u4f1a\u8bae\u65e5\u7a0b<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

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

\u853b\u4eab\u76f4\u64ad (\u53ef\u626b\u7801\u89c2\u770b\uff09\uff1a<\/strong><\/span><\/p>\n

<\/section>\n

<\/section>\n
2022\/12\/31<\/strong><\/span><\/section>\n

<\/p>\n

<\/section>\n
2023\/1\/1<\/strong><\/span><\/section>\n

<\/p>\n

<\/section>\n

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

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

\u62a5\u544a\u6458\u8981<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n
 <\/span><\/section>\n
\u5434\u82f1\u5e74\uff08UCLA\uff09<\/strong><\/span><\/section>\n
Representational Learning and Grid Cells<\/strong><\/span><\/section>\n

<\/section>\n
<\/section>\n
A key perspective of deep learning is representation learning, where concepts or entities are embedded in latent spaces and are represented by latent vectors whose elements can be interpreted as activities of neurons. In this talk, I will discuss our recent work on representational models of grid cells. The grid cells in the mammalian entorhinal cortex exhibit striking hexagon firing patterns when the agent (e.g., a rat or a human) navigates in the 2D open field. I will explain that the grid cells collectively form a vector representation of the 2D self-position, and the 2D self-motion is represented by the transformation of the vector. We identify a group representation condition and a conformal isometry condition for the transformation, and show that these two conditions lead to locally conformal embedding and the hexagon grid patterns. Joint work with Ruiqi Gao, Jianwen Xie, Wenhao Zhang, Xue-Xin Wei and Song-Chun Zhu.<\/span><\/section>\n

<\/section>\n
\u5468\u6d77\u519b \uff08\u4e2d\u79d1\u9662\u7406\u8bba\u7269\u7406\u6240\uff09<\/strong><\/span><\/section>\n
\u4fa7\u5411\u9884\u6d4b\u7f16\u7801\u548c\u611f\u77e5\u5668\u5728\u7ebf\u4e3b\u52a8\u5b66\u4e60\u521d\u63a2<\/strong><\/span><\/section>\n

<\/section>\n
\u8fd1\u5e74\u6765\uff0c\u795e\u7ecf\u7f51\u7edc\u7684\u7edf\u8ba1\u7269\u7406\u7269\u7406\u5b66\u8d8a\u6765\u8d8a\u5f15\u8d77\u4eba\u4eec\u7684\u5173\u6ce8\u3002\u4ece\u81ea\u65cb\u73bb\u7483\u548c\u7ec4\u5408\u4f18\u5316\u7684\u89d2\u5ea6\u770b\u5f85\u795e\u7ecf\u7f51\u7edc\u7684\u53c2\u6570\u5b66\u4e60\u95ee\u9898\u662f\u4e00\u4e2a\u91cd\u8981\u89c6\u89d2\u3002\u795e\u7ecf\u7f51\u7edc\u7684\u5b66\u4e60\u8fc7\u7a0b\uff0c\u53ef\u4ee5\u770b\u6210\u662f\u5728\u7f51\u7edc\u4e2d\u5efa\u7acb\u4e00\u4e2a\u5185\u90e8\u6a21\u578b\u8868\u5f81\u5916\u90e8\u4e16\u754c\u7684\u8fc7\u7a0b\u3002\u6211\u5c06\u6c47\u62a5\u6211\u4eec\u5728\u4fa7\u5411\u9884\u6d4b\u7f16\u7801\u548c\u611f\u77e5\u5668\u5728\u7ebf\u4e3b\u52a8\u5b66\u4e60\u65b9\u9762\u7684\u4e00\u70b9\u70b9\u7406\u8bba\u548c\u5b9e\u8bc1\u7ed3\u679c\uff0c\u91cd\u70b9\u5728\u4fa7\u5411\u9884\u6d4b\u7f16\u7801\u3002\u5bf9\u4e8e\u4fa7\u5411\u9884\u6d4b\u7f16\u7801\u95ee\u9898\uff0c\u6211\u4e5f\u5c06\u5c55\u793a\u6700\u4f18\u9884\u6d4b\u6743\u91cd\u77e9\u9635\u7684\u5bf9\u79f0\u6027\u53d1\u751f\u6539\u53d8\u7684\u4e00\u4e9b\u521d\u6b65\u6570\u503c\u8ba1\u7b97\u7ed3\u679c\u3002<\/span><\/section>\n

<\/section>\n
\u8c22\u79b9 \uff08MIT\uff09<\/strong><\/span><\/section>\n
Human-like capacity limits in working memory models result from naturalistic sensory constraints<\/strong><\/span><\/section>\n

<\/section>\n
Working memory (WM) allows us to hold information temporarily and make complex decisions beyond reflexive response to stimuli. One prominent feature of WM is its capacity limit. Despite decades of study, the root of this limit is still not well-understood. Most previous accounts for this limit assume various forms of memory constraints and make strong, often oversimplified, assumptions about sensory representations of stimuli. In this work, we built visual-cognitive neural network models of WM that process raw sensory stimuli. In contrast to intuitions that capacity limit results from memory constraints, we found that pre-training the sensory region of our models with natural images poses enough constraints on models to exhibit human-like behavior patterns across a wide range of WM capacity tasks. In change detection tasks, the detection accuracy decreases rapidly when the number of stimuli to be remembered increases. In continuous report tasks, the fidelity of the report again decreases rapidly when more stimuli are shown. In contrast, models without realistic constraints on the sensory regions produce super-human performance in these tasks. Human-like behavior cannot be restored simply by restricting the size of these models or adding processing noise. Unlike phenomenological accounts of WM capacity, our neural network models allow us to test the neural mechanisms of capacity limitation. We found that the average neural activation in our model increases and then plateaus when more stimuli are presented, and capacity limitation appears to arise in a bottom-up fashion; both are broadly consistent with previous fMRI and electrophysiological studies. Our work suggests that many phenomena about WM capacity can be explained by sensory constraints. Our models offer a fresh perspective in our understanding of the origin of the WM capacity limit and highlight the importance of building models with realistic sensory processing even when studying memory and other high-level cognitive phenomena.<\/span><\/section>\n

<\/section>\n
\u8def\u5b50\u7ae5 \uff08Ohio State University \uff09<\/strong><\/span><\/section>\n
Facial representation comparisons between human brain and hierarchical deep convolutional neural network reveal a fatigue repetition suppression mechanism.<\/strong><\/span><\/section>\n

<\/section>\n
Repetition suppression (RS) for faces, a phenomenon that neural responses are reduced to repeated faces in the visual cortex, have long been studied. However, the underlying primary neural mechanisms of RS remains debated. In recent years, artificial neural networks can achieve the performance of face recognition at human level. In our current study, we combined human electroencephalogram (EEG) and a hierarchical deep convolutional neural network (DCNN) and applied reverse engineering to provide a novel way to investigate the neural mechanisms of facial RS. First, we used brain decoding approach to explore the representations of faces and demonstrates its repetition suppression effect in human brains. Then not only we investigated how facial information was encoded in DCNN, but also we used DCNN as a tool to simulate the neural machanism of facial repetition suppression as fatigue or sharpening of neurons and compared representations between human brains and DCNNs. In fatigue hypothesis, we assumed that the activation of neurons with stronger response to face stimulus would be attenuated under the repetition condition. In sharpening hypothesis, we assumed that the neurons with weak response to face stimulus would be not activated any more under the repetition condition. As a core part, we constructed two RS models, fatigue and sharpening models, to modify the activations of DCNNs and conducted cross-modal representational similarity analysis (RSA) comparisons between dynamic processing in human EEG signals and layers in modified DCNNs respectively. We found that representations of human brains were more similar to those in fatigue-modified DCNN, compared with sharpening modified DCNN. Also, human brains showed stronger and longer similarities with the fatigue-modified DCNN as the layer increased. Our results suggests that the facial RS effect in face perception is more likely caused by the fatigue mechanism suggesting that the activation of neurons with stronger responses to face stimulus would be attenuated more. Therefore, the current study supports the fatigue mechanism as a more plausible neural mechanism of facial RS. The comparison between representations in the human brain and hierarchical DCNN provides a promising tool to simulate and infer the brain mechanism underlying human behaviors.<\/span><\/section>\n

<\/section>\n
\u6bd5\u5b8f\u6770 (\u6df1\u5733\u6e7e\u5b9e\u9a8c\u5ba4\u795e\u7ecf\u75be\u75c5\u7814\u7a76\u6240\uff09<\/strong><\/span><\/section>\n
\u4e0b\u4e00\u4ee3\u795e\u7ecf\u8d28\u91cf\u6a21\u578b\u7684\u7814\u7a76\u4e0e\u5e94\u7528<\/strong><\/span><\/section>\n

<\/section>\n
\u6458\u8981\uff1a\u672c\u62a5\u544a\u9996\u5148\u7b80\u8981\u6982\u8ff0\u4e86\u4e0b\u4e00\u4ee3\u795e\u7ecf\u8d28\u91cf\u6a21\u578b\uff0c\u5b83\u4ee3\u8868\u4e86\u5f02\u6784\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u7cbe\u786e\u5e73\u5747\u573a\u6a21\u578b\u53d1\u5c55\u7684\u65b0\u89c6\u89d2\u3002\u7136\u540e\u6211\u5c06\u62a5\u544a\u5e94\u7528\u8fd9\u79cd\u65b9\u6cd5\u6765\u91cd\u73b0\u795e\u7ecf\u79d1\u5b66\u4e2d\u89c2\u5bdf\u5230\u7684\u76f8\u5173\u5b9e\u9a8c\u73b0\u8c61\u7684\u7ed3\u679c\uff0c\u8303\u56f4\u4ece\u6162\u901f\u548c\u5feb\u901f\u4f3d\u9a6c\u632f\u8361\u5230\u5174\u594b\u6027\u795e\u7ecf\u5143\u9a71\u52a8\u7684\u7c7b\u4f3d\u9a6c\u632f\u8361\u4ee5\u53catheta\u632f\u8361\u5d4c\u5957\u7684\u4f3d\u9a6c\u632f\u8361\u3002\u6700\u540e\uff0c\u6211\u5c06\u5c55\u793a\u8fd9\u4e9b\u795e\u7ecf\u8d28\u91cf\u6a21\u578b\u5982\u4f55\u6269\u5c55\u4ee5\u6355\u83b7\u7531\u65e0\u5e8f\u7684\u52a8\u529b\u6e90\u5f15\u8d77\u7684\u6270\u52a8\u9a71\u52a8\u73b0\u8c61\uff0c\u8fd9\u4e9b\u6270\u52a8\u81ea\u7136\u5b58\u5728\u4e8e\u5927\u8111\u56de\u8def\u4e2d\uff0c\u4f8b\u5982\u80cc\u666f\u566a\u58f0\u6216\u8005\u7531\u4e8e\u795e\u7ecf\u5143\u4e4b\u95f4\u7684\u7a00\u758f\u8fde\u63a5\u800c\u5bfc\u81f4\u7684\u7535\u6d41\u6270\u52a8\u3002<\/span><\/section>\n

<\/section>\n
\u99ae\u5fd7\u8070 \uff08\u9999\u6e2f\u57ce\u5e02\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u6210\u9ad4\u795e\u7d93\u65b0\u751f\u5c0d\u6a21\u5f0f\u5206\u96e2\u7684\u589e\u76ca<\/strong><\/span><\/section>\n

<\/section>\n
Classically, neurogenesis (birth of neurons) was impossible for adult mammals. Since the second half of the last century, evidence has suggested that adult-born neurons exist in the dentate gyrus (DG) and the olfactory system. Suppressing adult neurogenesis in DG can result in the impairment of discriminating similar memories but not very-different memories. Also, studies found that eliminating adult neurogenesis may link to mood disorders. These results indicate that adult neurogenesis plays a vital role in neural processing. However, how adult neurogenesis contributes to neural information processing remains open. In our recent work, synaptic competition, a process taking part in adult neurogenesis, is essential for pattern separation. Also, we have designed an unsupervised learning rule based on synaptic competition. The learning rule outperforms back-propagation in some classification tasks, e.g., distinguishing digits from the MNIST dataset. Our results suggest that synaptic competition is the key to pattern separation and that competition-based learning could be helpful in machine learning.<\/span><\/section>\n

<\/section>\n
\u9b4f\u4f9d\u5a1c \uff08\u4e4b\u6c5f\u5b9e\u9a8c\u5ba4\uff09<\/strong><\/span><\/section>\n
\u7761\u7720\u5bf9\u8bb0\u5fc6\u5de9\u56fa\u7684\u4f5c\u7528\u53ca\u673a\u5236\u7814\u7a76<\/strong><\/span><\/section>\n

<\/section>\n
\u6bcf\u4e2a\u4eba\u5927\u7ea6\u82b1\u4e09\u5206\u4e4b\u4e00\u5de6\u53f3\u7684\u65f6\u95f4\u5728\u7761\u7720\u4e0a\u3002\u7761\u7720\u8fc7\u7a0b\u4e2d\u867d\u7136\u5927\u8111\u4e0e\u5916\u754c\u51e0\u4e4e\u5207\u65ad\u4e86\u8054\u7cfb\uff0c\u4f46\u662f\u5927\u8111\u4e2d\u7684\u795e\u7ecf\u7ec6\u80de\u5374\u5fd9\u788c\u5730\u91cd\u653e\u7740\u6e05\u9192\u72b6\u6001\u4e0b\u5b66\u4e60\u5230\u7684\u4fe1\u606f\uff0c\u4ece\u800c\u5b9e\u73b0\u8bb0\u5fc6\u7684\u5de9\u56fa\u3002\u7136\u800c\uff0c\u7761\u7720\u5bf9\u8bb0\u5fc6\u5de9\u56fa\u7684\u4f5c\u7528\u673a\u7406\u5c1a\u4e0d\u660e\u786e\uff0c\u800c\u4e14\u5f88\u96be\u5728\u6d3b\u4f53\u4e2d\u8fdb\u884c\u795e\u7ecf\u751f\u7269\u5b66\u673a\u5236\u7684\u7814\u7a76\u3002\u672c\u8bfe\u9898\u7ec4\u63d0\u51fa\u901a\u8fc7\u6784\u5efa\u5177\u6709\u751f\u7269\u7269\u7406\u5b66\u7279\u6027\u7684\u4e18\u8111-\u76ae\u5c42\u7f51\u7edc\u8ba1\u7b97\u6a21\u578b\uff0c\u63a2\u7d22\u4e86\u4e0d\u540c\u795e\u7ecf\u9012\u8d28\u5bf9\u975e\u5feb\u52a8\u773c\u7761\u7720\u4e2d\u7684\u8111\u7535\u4fe1\u53f7(\u6bd4\u5982\u7eba\u9524\u6ce2\uff0c\u6162\u6ce2\uff0c\u7eba\u9524\u6ce2\u6162\u6ce2\u8026\u5408\u7b49)\u7684\u4ea7\u751f\uff0c\u4ee5\u53ca\u5176\u5728\u8bb0\u5fc6\u5de9\u56fa\u4efb\u52a1\u4e2d\u7684\u4f5c\u7528\u548c\u673a\u5236\u3002\u6b64\u7814\u7a76\u6709\u671b\u5c06\u7761\u7720\u4e2d\u7684\u8111\u7535\u4fe1\u53f7\u4f5c\u4e3a\u963f\u5c14\u5179\u6d77\u9ed8\u75c7\u7b49\u8ba4\u77e5\u529f\u80fd\u969c\u788d\u75be\u75c5\u7684\u751f\u7269\u6807\u5fd7\u7269\uff0c\u4e3a\u671f\u65e9\u671f\u9884\u9632\u3001\u8bca\u65ad\u53ca\u5e72\u9884\u63d0\u4f9b\u575a\u5b9e\u7684\u7406\u8bba\u4f9d\u636e\u3002<\/span><\/section>\n

<\/section>\n
\u4e01\u5065\u8c6a \uff08\u5317\u4eac\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u7aef\u5230\u7aef\u53cd\u5411\u4f20\u64ad\u662f\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u7684\u53cc\u5203\u5251\u5417<\/strong><\/span><\/section>\n

<\/strong><\/span><\/section>\n
\u6458\u8981\uff1a\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u5728\u8fd1\u5e74\u7531\u4e8e\u53d7\u5230\u4f4e\u80fd\u8017\u5e94\u7528\u9700\u6c42\u800c\u5907\u53d7\u5173\u6ce8\u3002\u7aef\u5230\u7aef\u53cd\u5411\u4f20\u64ad\u7b97\u6cd5\u7ed9\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u8bad\u7ec3\u5e26\u6765\u4e86\u673a\u4f1a\u548c\u6311\u6218\u3002\u672c\u6b21\u62a5\u544a\u5c06\u4ecb\u7ecd\u5927\u89c4\u6a21\u6df1\u5ea6\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u5982\u4f55\u8003\u8651\u5229\u7528\u65f6\u5e8f\u4fe1\u606f\u5206\u5e03\u8fdb\u884c\u7aef\u5230\u7aef\u5b66\u4e60\uff1b\u540c\u65f6\u7aef\u5230\u7aef\u53cd\u5411\u4f20\u64ad\u7b97\u6cd5\u4e5f\u7ed9\u7f51\u7edc\u5e26\u6765\u4e86\u9c81\u68d2\u6027\u6311\u6218\uff0c\u5c06\u4ecb\u7ecd\u5927\u89c4\u6a21\u6df1\u5ea6\u8109\u51b2\u795e\u7ecf\u7f51\u7edc\u5982\u4f55\u5e94\u5bf9\u6311\u6218\u5e76\u63d0\u9ad8\u9c81\u68d2\u6027\u3002<\/span><\/section>\n

<\/section>\n
\u848b\u5b50\u5065 \uff08\u4e2d\u5c71\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u975e\u5bf9\u79f0\u9ad8\u9636\u8054\u60f3\u8bb0\u5fc6\u6a21\u578b\u7684\u968f\u673a\u77e9\u9635\u7814\u7a76<\/strong><\/span><\/section>\n

<\/strong><\/span><\/section>\n
\u60c5\u666f\u8bb0\u5fc6\u53ea\u80fd\u88ab\u7f16\u7801\u4e8e\u5177\u6709\u975e\u5bf9\u79f0\u8fde\u63a5\u6743\u91cd\u7684\u795e\u7ecf\u7f51\u7edc\u4e2d\u3002\u5728\u6b64\u57fa\u7840\u4e0a\uff0c\u6211\u4eec\u7ed3\u5408\u4e86\u5b9e\u9a8c\u4e2d\u5e7f\u6cdb \u89c2\u5bdf\u5230\u7684\u7a81\u89e6\u53ef\u5851\u6027\u7684\u7075\u6d3b\u65f6\u95f4\u7a97\u53e3\uff0c\u63d0\u51fa\u4e86\u975e\u5bf9\u79f0\u7684\u9ad8\u9636\u8054\u60f3\u8bb0\u5fc6\u6a21\u578b\u3002\u7531\u4e8e\u975e\u5bf9\u79f0\u6743\u91cd\u5e26\u6765 \u7684\u975e\u5e73\u8861\u6001\u52a8\u529b\u5b66\uff0c\u7406\u8bba\u7814\u7a76\u76ee\u524d\u53ea\u6709\u5f88\u6709\u9650\u7684\u624b\u6bb5\u3002\u6211\u4eec\u4ece\u5c40\u57df\u52a8\u529b\u5b66\u7684\u89d2\u5ea6\u8fdb\u884c\u4e86\u5207\u5165\u2014\u2014 \u8fd0\u7528\u968f\u673a\u77e9\u9635\u5de5\u5177\u8ba1\u7b97\u6a21\u578b\u52a8\u529b\u5b66\u7684\u96c5\u53ef\u6bd4\u77e9\u9635\u7684\u672c\u5f81\u503c\u8c31\uff0c\u4ece\u4e2d\u83b7\u5f97\u5c40\u57df\u52a8\u529b\u5b66\u7684\u5173\u952e\u6027\u8d28\uff0c \u5e76\u4e0e\u5168\u5c40\u52a8\u529b\u5b66\u76f8\u4e92\u8054\u7cfb\uff0c\u6700\u7ec8\u9884\u6d4b\u60c5\u666f\u8bb0\u5fc6\u7684\u8fd0\u884c\u3002\u4e0d\u6b62\u4e8e\u6b64\uff0c\u8fd9\u5957\u57fa\u4e8e\u968f\u673a\u77e9\u9635\u5de5\u5177\u7684\u7814\u7a76 \u65b9\u6cd5\u4e5f\u5c06\u9002\u7528\u4e8e\u8bb8\u591a\u5176\u4ed6\u795e\u7ecf\u7f51\u7edc\u573a\u666f\u3002<\/span><\/section>\n

<\/section>\n
\u7530\u4eae \uff08\u9999\u6e2f\u6d78\u4f1a\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u4ece\u4eba\u5de5\u795e\u7ecf\u7f51\u7edc\u7684\u89c6\u89d2\u523b\u753b\u4e2d\u533b\u7406\u8bba<\/strong><\/span><\/section>\n

<\/section>\n
Traditional Chinese Medicine (TCM), which originated in ancient China with a history of thousands of years, characterizes and addresses human physiology, pathology, and diseases diagnosis and prevention using TCM theories and Chinese herbal products. However, the pharmacological principles in TCM theory, the core treasure house of TCM, have rarely been systematically investigated in a top-down manner, which hinders the modernization and standardization of TCM. In this work, we proposed a novel TCM-based network pharmacology framework to discern general patterns and principles of human disease and predict herb-diseases associations. Specifically, we constructed an integrative database and a pharmacological network of TCM through extensively collecting and cleaning large-scale TCM prescription data from ancient books, modern literature, and existing TCM data resources. Various topological and structural properties of the TCM pharmacological network were systematically characterized to decipher the pharmacological principles of TCM theory. Based on the TCM pharmacological network, we uncovered the human disease-disease relationship and build an in-silico network-based pipeline for the prediction of drug-disease associations. Our work promotes the quantitative underpinning of TCM pharmacological principles, provides a basis for the objectification of the diagnosis and treatment process of Chinese medicine, and paves the way for the knowledge fusion of TCM evidence-based medicine and modern biology.<\/span><\/section>\n

<\/section>\n
\u96f7\u5a1c \uff08\u5927\u8fde\u7406\u5de5\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
An Optimal Transport View of Deep Learning<\/strong><\/span><\/section>\n

<\/section>\n
Abstract: In this talk we first introduce an optimal transportation (OT) view of deep learning. Natural datasets have intrinsic patterns, which can be summarized as the manifold distribution principle: the distribution of a class of data is close to a low-dimensional manifold. Deep learning mainly accomplish two tasks: manifold learning and probability distribution transformation. The latter can be carried out using the classical OT method. From the OT perspective, the generator of GAN model computes the OT map, while the discriminator computes the Wasserstein distance between the generated data distribution and the real data distribution; both can be reduced to a convex geometric optimization process. Furthermore, OT theory discovers the intrinsic collaborative\u2014instead of competitive\u2014relation between the generator and the discriminator. Then we give an explanation to mode collapse. By the regularity theory of Monge-Ampere equation the OT map is discontinued if the support of the target distribution is non-convex. But the neural network always represents continue map. This is the reason of mode collapse. Finally based on the above theory we propose three novel models, including generative model AE-OT, super resolutin model OTSR and point cloud upsampling model PU-CycGAN.<\/span><\/section>\n

<\/section>\n
\u8463\u884c\u601d \uff08\u5317\u4eac\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
Adaptation Accelerating Sampling-based Bayesian Inference in Attractor Neural Networks<\/strong><\/span><\/section>\n
\u8d1f\u53cd\u9988\u5e2e\u52a9\u52a0\u901f\u5438\u5f15\u5b50\u7f51\u7edc\u7684\u91c7\u6837<\/strong><\/span><\/section>\n

<\/section>\n
The brain performs probabilistic Bayesian inference to interpret the external world. The sampling-based view assumes that the brain represents the stimulus posterior distribution via samples of stochastic neuronal responses. Although the idea of sampling-based inference is appealing, it faces a critical challenge of whether stochastic sampling is fast enough to match the rapid computation of the brain. In this study, we explore how latent feature sampling can be accelerated in neural circuits. Specifically, we consider a canonical neural circuit model called con- tinuous attractor neural networks (CANNs) and investigate how sampling-based inference of latent continuous variables is accelerated in CANNs. Intriguingly, we find that by including noisy adaptation in the neuronal dynamics, the CANN is able to speed up the sampling process significantly. We theoretically derive that the CANN with noisy adaptation implements the efficient sampling method called Hamiltonian dynamics with friction, where noisy adaption effectively plays the role of momentum. We theoretically analyze the sampling performances of the network and derive the condition when the acceleration has the maximum effect. Simulation results validate our theoretical analyses. We further extend the model tocoupled CANNs and demonstrate that noisy adaptation accelerates the sampling of the posterior distribution of multivariate stimuli. We hope that this study enhances our understanding of how Bayesian inference is realized in the brain.<\/span><\/section>\n
\u5927\u8111\u9700\u8981\u6267\u884c\u5927\u91cf\u7684\u6982\u7387\u63a8\u65ad\u4efb\u52a1\uff0c\u91c7\u6837\u5b66\u6d3e\u8ba4\u4e3a\u795e\u7ecf\u5143\u7684\u968f\u673a\u53d1\u653e\u662f\u5728\u5bf9\u6240\u9700\u63a8\u65ad\u7684\u6982\u7387\u8fdb\u884c\u91c7\u6837\uff0c\u4f46\u662f\u4f20\u7edf\u7684\u91c7\u6837\u7b97\u6cd5\u8f83\u6162\uff0c\u5f88\u96be\u540c\u5927\u8111\u7684\u5feb\u901f\u53cd\u5e94\u5bf9\u5e94\u8d77\u6765\uff0c\u6211\u4eec\u7684\u5de5\u4f5c\u63d0\u51fa\u4e86\u4e00\u79cd\u5728\u8fde\u7eed\u5438\u5f15\u5b50\u7f51\u8def\u91cc\u5b9e\u73b0\u5feb\u901f\u91c7\u6837\u7b97\u6cd5\u7684\u673a\u5236\uff0c\u7f51\u7edc\u6ce2\u5305\u5728\u91c7\u6837\u7684\u8fc7\u7a0b\u4e2d\u5f15\u5165\u8d1f\u53cd\u9988\u76f8\u5f53\u4e8e\u7ed9\u5176\u589e\u52a0\u4e86\u52a8\u91cf\u9879\uff0c\u4f7f\u5176\u53ef\u4ee5\u7d2f\u79ef\u8fc7\u5f80\u7684\u901f\u5ea6\uff0c\u4ece\u800c\u63d0\u9ad8\u91c7\u6837\u901f\u5ea6\uff0c\u6211\u4eec\u7406\u8bba\u8bc1\u660e\u4e86\u8fd9\u79cd\u673a\u5236\u5bf9\u5e94\u4e86\u9ad8\u6548\u7684\u54c8\u5bc6\u987f\u91c7\u6837\u7b97\u6cd5\uff1b\u540c\u65f6\u6211\u4eec\u8bc1\u660e\u5728\u591a\u4e2a\u8026\u5408\u7684\u5438\u5f15\u5b50\u7f51\u7edc\u4e2d\u53ef\u4ee5\u5b9e\u73b0\u5bf9\u591a\u53d8\u91cf\u7684\u5206\u5e03\u5f0f\u54c8\u5bc6\u987f\u91c7\u6837\uff0c\u6211\u4eec\u7684\u5b9e\u9a8c\u7ed3\u679c\u4e5f\u4e0e\u7406\u8bba\u7ed3\u679c\u76f8\u543b\u5408\u3002<\/span><\/section>\n

<\/section>\n
\u90b9\u6587\u8f69\uff08\u4e2d\u5c71\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u8fde\u7eed\u5b66\u4e60\u7684\u7edf\u8ba1\u529b\u5b66 \u2014\u2014 \u4ece\u6a21\u578b\u5230\u7b97\u6cd5<\/strong><\/span><\/section>\n

<\/section>\n
\u707e\u96be\u6027\u9057\u5fd8\u76ee\u524d\u4ecd\u7136\u662f\u6df1\u5ea6\u5b66\u4e60\u4e9f\u5f85\u89e3\u51b3\u7684\u96be\u9898\u4e4b\u4e00\uff0c\u5373\u5982\u4f55\u8ba9\u7b97\u6cd5\u5728\u5b66\u4e60\u65b0\u4efb\u52a1\u7684\u540c\u65f6\uff0c\u7ef4\u6301\u5386\u53f2\u4efb\u52a1\u7684\u8868\u73b0\u4ece\u800c\u5b9e\u73b0\u8fde\u7eed\u5b66\u4e60\u3002\u4e0d\u5c11\u6765\u81ea\u673a\u5668\u5b66\u4e60\u754c\u53ca\u8ba1\u7b97\u795e\u7ecf\u79d1\u5b66\u754c\u7684\u542f\u53d1\u5f0f\u6280\u5de7\uff0c\u90fd\u80fd\u4e00\u5b9a\u7a0b\u5ea6\u4e0a\u7f13\u89e3\u9057\u5fd8\u3002\u8fd9\u91cc\uff0c\u5173\u4e8e\u4e8c\u503c\u795e\u7ecf\u7f51\u7edc\uff0c\u6211\u4eec\u63d0\u51fa\u4e86\u4e00\u5957\u7edf\u4e00\u7684\u5b66\u4e60\u6846\u67b6\uff1a\u91c7\u7528\u53d8\u5206\u6846\u67b6\u5c06\u7f51\u7edc\u7684\u8bad\u7ec3\u4e3b\u4f53\u6539\u4e3a\u6743\u91cd\u5747\u503c\uff0c\u518d\u901a\u8fc7\u7b80\u5355\u7684\u7ebf\u6027\u54cd\u5e94\u5173\u7cfb\uff0c\u5728\u5916\u573a\u7a7a\u95f4\u4e2d\u8fdb\u884c\u68af\u5ea6\u4e0b\u964d\uff1b\u7531\u6b64\uff0c\u4e0d\u786e\u5b9a\u6027\u5c06\u81ea\u7136\u5730\u88ab\u5f15\u5165\u53d8\u5206\u6846\u67b6\u4e2d\uff0c\u5728\u8fde\u7eed\u5b66\u4e60\u7684\u8fc7\u7a0b\u4e2d\u8d77\u5230\u91cd\u8981\u4f5c\u7528\u3002\u57fa\u4e8e\u6b64\u5b66\u4e60\u6846\u67b6\uff0c\u6211\u4eec\u9996\u5148\u4ece\u7edf\u8ba1\u529b\u5b66\u7684\u89d2\u5ea6\uff0c\u5728\u5355\u5c42\u7f51\u7edc\u4e2d\u8fdb\u884c\u5206\u6790\uff1b\u5177\u4f53\u6765\u8bf4\uff0c\u6211\u4eec\u501f\u7528\u4e86Franz-Parisi\u71b5\u7684\u6982\u5ff5\u6784\u5efa\u4e86\u7f51\u7edc\u7684\u52bf\u51fd\u6570\uff0c\u5e76\u4f7f\u7528\u590d\u672c\u65b9\u6cd5\u6c42\u89e3\uff0c\u7531\u6b64\u5f97\u5230\u7684\u5e8f\u53c2\u91cf\u5f88\u597d\u5730\u63cf\u8ff0\u4e86\u5b66\u4e60\u7684\u8fc7\u7a0b\uff0c\u5e76\u80fd\u9884\u6d4b\u8fde\u7eed\u5b66\u4e60\u7684\u8bef\u5dee\u3002\u968f\u540e\uff0c\u6211\u4eec\u5728\u6df1\u5ea6\u7f51\u7edc\u4e2d\u8fd0\u7528\u6b64\u5b66\u4e60\u6846\u67b6\u8bbe\u8ba1\u7b97\u6cd5\uff0c\u5e76\u5728\u6d89\u53ca\u771f\u5b9e\u6570\u636e\u7684\u8fde\u7eed\u5b66\u4e60\u4efb\u52a1\u4e0a\u8fdb\u884c\u6d4b\u8bd5\uff0c\u8868\u73b0\u51fa\u975e\u5e38\u60ca\u4eba\u7684\u8fde\u7eed\u5b66\u4e60\u6548\u679c\u3002\u6700\u540e\uff0c\u6211\u4eec\u63d0\u51fa\u7684\u5b66\u4e60\u6846\u67b6\u8fd8\u4e0e\u673a\u5668\u5b66\u4e60\u4e2d\u7ecf\u5178\u7684EWC(elastic weight consolidation)\u7b97\u6cd5\uff0c\u4ee5\u53ca\u8ba1\u7b97\u795e\u7ecf\u79d1\u5b66\u4e2d\u7684\u6743\u91cd\u518d\u53ef\u5851\u6027(meta plasticity)\u76f8\u5173\uff1b\u56e0\u6b64\u6211\u4eec\u7684\u5b66\u4e60\u6846\u67b6\u8bc1\u660e\u4e86\u57fa\u4e8e\u7edf\u8ba1\u7269\u7406\u6a21\u578b\u8bbe\u8ba1\u4eba\u5de5\u667a\u80fd\u8fde\u7eed\u5b66\u4e60\u7b97\u6cd5\u7684\u6709\u6548\u6027\u3002<\/span><\/section>\n

<\/section>\n
\u5b59\u6d69\u817e \uff08\u5409\u6797\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
Image encryption based on hyperchaos combined with different methods<\/strong><\/span><\/section>\n

<\/section>\n
With the development of digital information technology, information security has been widely concerned, and multimedia encryption technology is one of the important technologies to ensure information security. Chaotic sequence cryptography has become an important branch of cryptography as a hot research topic in the field of cryptography. With hyperchaotic system combined with methods like DNA coding and neural network, the image encryption algorithm has the features of complex structure, large key space and strong resistance to attack, which can effectively meet the characteristics of multimedia encryption, so it has a good prospect to be applied to multimedia encryption.<\/span><\/section>\n

<\/section>\n
\u90d1\u96c5\u83c1 \uff08\u5317\u4eac\u5927\u5b66\uff09<\/strong><\/span><\/section>\n
\u89c6\u7f51\u819c\u7684\u795e\u7ecf\u7f51\u7edc\u7f16\u7801\u6a21\u578b<\/strong><\/span><\/section>\n

<\/section>\n
\u5927\u8111\u4e2d\u6240\u63a5\u6536\u7684\u4fe1\u606f\u8d85\u8fc770%\u90fd\u662f\u6765\u81ea\u4e8e\u89c6\u89c9\u7cfb\u7edf\uff0c\u800c\u89c6\u7f51\u819c\u4f5c\u4e3a\u5fc3\u7075\u4e4b\u7a97\uff0c\u662f\u751f\u7269\u89c6\u89c9\u4fe1\u606f\u5904\u7406\u7684\u7b2c\u4e00\u7ad9\uff0c\u8d1f\u8d23\u5bf9\u65f6\u7a7a\u4e2d\u4e0d\u65ad\u53d8\u5316\u7684\u53ef\u89c1\u5149\u8fdb\u884c\u5b9e\u65f6\u7f16\u7801\u3002\u63a2\u7d22\u89c6\u7f51\u819c\u7684\u7f16\u7801\u673a\u5236\u5bf9\u4e8e\u63ed\u793a\u5176\u5b83\u89c6\u89c9\u7cfb\u7edf\u7684\u8ba1\u7b97\u539f\u7406\u81f3\u5173\u91cd\u8981\u3002\u800c\u968f\u7740\u6df1\u5ea6\u5b66\u4e60\u7684\u53d1\u5c55\uff0c\u4eba\u5de5\u795e\u7ecf\u7f51\u7edc\u5df2\u5728\u8ba1\u7b97\u673a\u89c6\u89c9\u3001\u81ea\u7136\u8bed\u8a00\u5904\u7406\u7b49\u9886\u57df\u53d6\u5f97\u4f18\u8d8a\u7684\u6027\u80fd\u3002\u4f7f\u7528\u4eba\u5de5\u795e\u7ecf\u7f51\u7edc\u5b66\u4e60\u89c6\u7f51\u819c\u6216\u8005\u521d\u7ea7\u89c6\u76ae\u5c42\u54cd\u5e94\u4e0e\u8f93\u5165\u523a\u6fc0\u7684\u6620\u5c04\u5173\u7cfb\uff0c\u5bf9\u4e86\u89e3\u89c6\u7f51\u819c\u7b49\u521d\u7ea7\u89c6\u89c9\u7cfb\u7edf\u7684\u5de5\u4f5c\u673a\u5236\u5177\u6709\u5de8\u5927\u4f18\u52bf\u3002\u89c6\u7f51\u819c\u7f16\u7801\u6a21\u578b\u7684\u7814\u7a76\u4e0d\u4ec5\u5177\u6709\u91cd\u8981\u7684\u7406\u8bba\u7814\u7a76\u610f\u4e49\uff0c\u800c\u4e14\u5177\u6709\u5b9e\u9645\u7684\u5de5\u7a0b\u610f\u4e49\u3002\u901a\u8fc7\u63ed\u793a\u89c6\u7f51\u819c\u52a0\u5de5\u5904\u7406\u4fe1\u606f\u7684\u5de5\u4f5c\u673a\u7406\uff0c\u4e00\u65b9\u9762\u53ef\u4ee5\u4e3a\u89c6\u89c9\u795e\u7ecf\u5047\u4f53\u7684\u7814\u5236\u63d0\u4f9b\u7406\u8bba\u57fa\u7840\uff1b\u53e6\u4e00\u65b9\u9762\u53ef\u4ee5\u542f\u53d1\u8bbe\u8ba1\u66f4\u52a0\u667a\u80fd\u7684\u795e\u7ecf\u5f62\u6001\u89c6\u89c9\u4f20\u611f\u5668\uff0c\u4f8b\u5982\u4e8b\u4ef6\u76f8\u673a\u4e0e\u8109\u51b2\u76f8\u673a\u3002\u672c\u62a5\u544a\u5c06\u7740\u91cd\u4ecb\u7ecd\u57fa\u4e8e\u795e\u7ecf\u7f51\u7edc\u7684\u5355\u795e\u7ecf\u8282\u7ec6\u80de\u53ca\u7fa4\u4f53\u795e\u7ecf\u8282\u7ec6\u80de\u7f16\u7801\u6a21\u578b\u3002\u8fd9\u4e9b\u6a21\u578b\u80fd\u5728\u5b66\u4e60\u89c6\u89c9\u523a\u6fc0\u4e0e\u795e\u7ecf\u8282\u7ec6\u80de\u54cd\u5e94\u6620\u5c04\u5173\u7cfb\u7684\u540c\u65f6\uff0c\u63ed\u793a\u6a21\u62df\u5355\u4e2a\u795e\u7ecf\u8282\u795e\u7ecf\u5143\u529f\u80fd\u7684\u6700\u5c0f\u795e\u7ecf\u7f51\u7edc\u7cfb\u7edf\u53ca\u7fa4\u4f53\u7ec6\u80de\u7f16\u7801\u52a8\u6001\u523a\u6fc0\u7684\u5173\u952e\u56e0\u7d20\u3002<\/span><\/section>\n

<\/section>\n
\u7941\u9633 \uff08\u590d\u65e6\u5927\u5b66)<\/strong><\/span><\/section>\n
Fractional neural sampling: spatiotemporal probabilistic computations in spiking neural circuits<\/strong><\/span><\/section>\n

<\/section>\n
Neurophysiological recordings have revealed large fluctuations of spiking activity of cortical neurons both over time and across trials, leading to the idea that neural computing in the brain is fundamentally probabilistic. A range of perceptual and cognitive processes have been characterized from the perspective of probabilistic representations and inference. To understand the neural circuit mechanism underlying these probabilistic computations, we develop a theory based on complex spatiotemporal dynamics of neural population activity. By employing a biophysically realistic spiking neural circuit, we show the emergence of spatially structured population activity patterns capturing realistic variability of neural dynamics both in time and in space. These activity patterns exhibit large jumps with fractional order statistics and implement a type of probabilistic computations that we name fractional neural sampling (FNS). We further develop a mathematical model to reveal the algorithmic nature of FNS and its computational advantages for representing multimodal distributions. The FNS theory thus provides a unified account of a diversity of experimental observations of neural spatiotemporal dynamics and perceptual processes, and further provide links between the structure, dynamics, and function of cortical circuits.<\/span><\/section>\n

<\/section>\n
\u8bb8\u5fd7\u94a6 \uff08\u4e0a\u6d77\u4ea4\u5927\uff09<\/strong><\/span><\/section>\n
\u795e\u7ecf\u7f51\u7edc\u7684\u51dd\u805a\u73b0\u8c61<\/strong><\/span><\/section>\n

<\/section>\n
\u4e3a\u4ec0\u4e48\u770b\u8d77\u6765\u5982\u6b64\u590d\u6742\u7684\u795e\u7ecf\u7f51\u7edc(NN)\u901a\u5e38\u6cdb\u5316\u5f97\u5f88\u597d?\u4e3a\u4e86\u7406\u89e3\u8fd9\u4e2a\u95ee\u9898\uff0c\u6211\u4eec\u5728\u8bad\u7ec3\u795e\u7ecf\u7f51\u7edc\u65f6\u53d1\u73b0\u4e86\u4e00\u4e9b\u7b80\u5355\u7684\u9690\u5f0f\u6b63\u5219\u5316\u3002\u9996\u5148\u662f\u795e\u7ecf\u7f51\u7edc\u4ece\u4f4e\u9891\u5b66\u4e60\u5230\u9ad8\u9891\u7684\u9891\u7387\u539f\u7406\uff08\u9891\u7387\u539f\u5219\uff09\u3002\u5176\u6b21\u662f\u53c2\u6570\u51dd\u805a\uff0c\u8fd9\u662f\u975e\u7ebf\u6027\u8bad\u7ec3\u8fc7\u7a0b\u7684\u4e00\u4e2a\u7279\u70b9\uff0c\u5b83\u6709\u6548\u5730\u51cf\u5c0f\u4e86\u7f51\u7edc\u7684\u89c4\u6a21\u3002\u5728\u6b64\u57fa\u7840\u4e0a\uff0c\u6211\u4eec\u53d1\u73b0\u4e86\u795e\u7ecf\u7f51\u7edc\u635f\u5931\u666f\u89c2\u7684\u5185\u5728\u5d4c\u5165\u539f\u7406\u4ee5\u53ca\u7406\u89e3\u5927\u7f51\u7edc\u5728\u8bad\u7ec3\u4e0a\u5e26\u6765\u4e00\u4e9b\u4f18\u52bf\uff0c\u5e76\u53d1\u5c55\u4e86\u4e00\u4e2a\u79e9\u5206\u6790\u6846\u67b6\uff0c\u5b9a\u91cf\u5730\u7406\u89e3\u4e00\u4e2a\u8fc7\u5ea6\u53c2\u6570\u5316\u7684\u795e\u7ecf\u7f51\u7edc\u9700\u8981\u591a\u5c11\u6570\u636e\u91cf\uff0c\u4ee5\u4fbf\u66f4\u597d\u5730\u6cdb\u5316\u3002<\/span><\/section>\n

<\/section>\n

<\/section>\n

\n
\n
\n
\n
\n

<\/section>\n

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

\u81f4\u8c22<\/span><\/strong><\/strong><\/p>\n<\/section>\n

\n

<\/section>\n

<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/section>\n<\/h3>\n
 <\/span><\/section>\n
\u56fd\u5bb6\u81ea\u7136\u79d1\u5b66\u57fa\u91d1No 12122515<\/span><\/section>\n

<\/mp-style-type><\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"

\u8bba\u575b\u7b80\u4ecb   \u6df1\u5ea6\u795e\u7ecf\u7f51\u7edc\u6b63\u5728\u6210\u4e3aAI for Science\u7684\u57fa\u672c\u5de5\u5177\uff0c\u6784\u6210\u65b0\u7684\u79d1\u5b66\u7814\u7a76\u8303\u5f0f\uff0c\u4ee5\u524d\u6240\u672a\u6709\u7684\u5e7f\u5ea6\u548c\u6df1\u5ea6\u5f71\u54cd\u79d1\u5b66\u7684\u53d1\u5c55\u3002\u7834\u89e3\u6df1\u5ea6\u7f51\u7edc\u7684\u673a\u5236\u4f7f\u5f97\u4eba\u4eec\u80fd\u4ece\u7b2c\u4e00\u6027\u539f\u7406\u51fa\u53d1\u8bbe\u8ba1\u66f4\u597d\u7684\u7f51\u7edc\u7ed3\u6784\u548c\u7b97\u6cd5\uff0c\u751a\u81f3\u6709\u52a9\u4e8e\u7406\u89e3\u751f\u7269\u667a\u80fd\u7684\u672c\u8d28\u3002\u795e\u7ecf\u7f51\u7edc\u65b0\u5e74\u4ea4\u53c9\u8bba\u575b\u4e8e2021\u5e74\u5e95\u7531\u4e00\u7fa4\u70ed\u8877\u4e8e\u4ea4\u53c9\u5b66\u79d1\u7814…<\/p>\n","protected":false},"author":1,"featured_media":39498,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[1],"tags":[],"special":[],"_links":{"self":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/39515"}],"collection":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=39515"}],"version-history":[{"count":0,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/posts\/39515\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=\/wp\/v2\/media\/39498"}],"wp:attachment":[{"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=39515"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=39515"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=39515"},{"taxonomy":"special","embeddable":true,"href":"https:\/\/swarma.org\/index.php?rest_route=%2Fwp%2Fv2%2Fspecial&post=39515"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}