- \n
- “Data-driven approach for functional brain image analysis
\n-Solution exploration of important functional connectivity using the Pareto solution set-”
\nA.FUJII; S.HIWA; T.HIROYASU.<\/li>\n<\/ul>\n<\/div>\n
\n\u3053\u306e\u30ef\u30fc\u30af\u30b7\u30e7\u30c3\u30d7\u306f\u3001\u9032\u5316\u8a08\u7b97\u5b66\u4f1a\u306b\u3088\u3063\u3066\u4e3b\u50ac\u3055\u308c\u305f\u56fd\u969b\u5b66\u4f1a\u3067\u3001\u9032\u5316\u8a08\u7b97\u5b66\u4f1a\u7814\u7a76\u4f1a\u3068\u3057\u3066\u306f14\u56de\u76ee\u3067\u521d\u306e\u6d77\u5916\u958b\u50ac\u3068\u306a\u308a\u307e\u3057\u305f\uff0e\u307e\u305f\u57fa\u8abf\u8b1b\u6f14\u3067\u306f\uff0c\u9999\u6e2f\u57ce\u5e02\u5927\u5b66\u306eQingfu Zhang\u6c0f\u3092\u306f\u3058\u3081\u8a08\u7b97\u79d1\u5b66\u306e\u5206\u91ce\u3092\u30ea\u30fc\u30c9\u3059\u308b\u5148\u751f\u65b9\u306b\u3088\u308b\u8b1b\u6f14\u304c\u884c\u308f\u308c\u307e\u3057\u305f\uff0e\u8b1b\u6f14\u306b\u3066\u9032\u5316\u8a08\u7b97\u3084\u6df1\u5c64\u5b66\u7fd2\u306e\u57fa\u672c\u7684\u306a\u77e5\u8b58\u304b\u3089\u6700\u5148\u7aef\u306e\u7814\u7a76\u307e\u3067\u805e\u304f\u3053\u3068\u304c\u3067\u304d\uff0c\u975e\u5e38\u306b\u6709\u610f\u7fa9\u306a\u6642\u9593\u3092\u904e\u3054\u3059\u3053\u3068\u304c\u3067\u304d\u307e\u3057\u305f\uff0e
\n\u4eca\u56de\u306e\u30ef\u30fc\u30af\u30b7\u30e7\u30c3\u30d7\u306f\u81ea\u8eab\u306e\u7814\u7a76\u304c\u8133\u6a5f\u80fd\u7814\u7a76\u3078\u3068\u5909\u308f\u3063\u3066\u304b\u3089\u521d\u306e\u5b66\u4f1a\u767a\u8868\u3067\u3042\u308a\uff0c\u521d\u306e\u6d77\u5916\u767a\u8868\u3067\u3057\u305f\uff0e\u305d\u306e\u305f\u3081\u4e0d\u5b89\u3082\u3042\u308a\u307e\u3057\u305f\u304c\uff0c\u30dd\u30b9\u30bf\u30fc\u30bb\u30c3\u30b7\u30e7\u30f3\u3067\u306f\u591a\u304f\u306e\u4ed6\u5927\u5b66\u306e\u5b66\u751f\u3084\u5148\u751f\u65b9\u306b\u8208\u5473\u3092\u6301\u3063\u3066\u3044\u305f\u3060\u304d\uff0c\u69d8\u3005\u306a\u610f\u898b\u3084\u8cea\u554f\u3092\u3044\u305f\u3060\u3044\u305f\u306e\u3067\uff0c\u3068\u3066\u3082\u5145\u5b9f\u3057\u305f\u6642\u9593\u3068\u306a\u308a\u307e\u3057\u305f\uff0e\u4eca\u56de\u306e\u767a\u8868\u3067\u5f97\u3089\u308c\u305f\u7d4c\u9a13\u3084\u81ea\u4fe1\u3092\u6b21\u56de\u4ee5\u964d\u306b\u3057\u3063\u304b\u308a\u3044\u304b\u3057\u3066\u3086\u304d\u305f\u3044\u3068\u601d\u3044\u307e\u3059\uff0e\u307e\u305f\u4eca\u56de\u306e\u30ef\u30fc\u30af\u30b7\u30e7\u30c3\u30d7\u3067\u5148\u751f\u65b9\u306e\u8b1b\u6f14\u3084\u5b66\u751f\u306e\u767a\u8868\u3092\u8074\u8b1b\u3057\uff0c\u3088\u308a\u826f\u3044\u30c7\u30a3\u30b9\u30ab\u30c3\u30b7\u30e7\u30f3\u3092\u3059\u308b\u4e0a\u3067\u81ea\u5206\u306e\u77e5\u8b58\u304c\u307e\u3060\u307e\u3060\u8db3\u308a\u306a\u3044\u3053\u3068\u3092\u5b9f\u611f\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3057\u305f\uff0e\u6b21\u56de\u306e\u5b66\u4f1a\u53c2\u52a0\u307e\u3067\u306b\u7814\u7a76\u3092\u901a\u3057\u3066\u3088\u308a\u591a\u304f\u306e\u77e5\u8b58\u3092\u8eab\u306b\u7740\u3051\u3066\u304a\u3053\u3046\u3068\u601d\u3044\u307e\u3059\uff0e
\n
\n![\"\"](\"http:\/\/www.is.doshisha.ac.jp\/news\/wp-content\/uploads\/2018\/08\/0119_xlarge-300x200.jpg\")
<\/a> ![\"\"](\"http:\/\/www.is.doshisha.ac.jp\/news\/wp-content\/uploads\/2018\/08\/0121_xlarge-258x300.jpg\")
<\/a>
\n
\n\u3010\u6587\u8cac\uff1aM1 \u85e4\u4e95\u3011
\n
\n
\n\u5b66\u4f1a\u53c2\u52a0\u5831\u544a\u66f8<\/strong><\/p>\n\n\n
\n \u5831\u544a\u8005\u6c0f\u540d<\/strong><\/td>\n \u85e4\u4e95 \u5149\u592e<\/td>\n<\/tr>\n \n \u767a\u8868\u8ad6\u6587\u30bf\u30a4\u30c8\u30eb<\/strong><\/td>\n Data-driven approach for functional brain image analysis<\/td>\n<\/tr>\n \n \u8457\u8005<\/strong><\/td>\n \u85e4\u4e95\u5149\u592e, \u65e5\u548c\u609f, \u5ee3\u5b89\u77e5\u4e4b<\/td>\n<\/tr>\n \n \u4e3b\u50ac<\/strong><\/td>\n \u9032\u5316\u8a08\u7b97\u5b66\u4f1a<\/td>\n<\/tr>\n \n \u5b66\u4f1a\u540d<\/strong><\/td>\n 2018 JPNSEC International Workshop on Evolutionary Computation<\/td>\n<\/tr>\n \n \u4f1a\u5834<\/strong><\/td>\n Southern University of Science and Technology (SUSTech), Shenzhen, China<\/td>\n<\/tr>\n \n \u958b\u50ac\u65e5\u7a0b<\/strong><\/td>\n 2018\/08\/31-2018\/9\/1<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
\n <\/p>\n- \n
- \u8b1b\u6f14\u4f1a\u306e\u8a73\u7d30<\/li>\n<\/ol>\n
2018\/08\/31\u304b\u30892018\/09\/1\u306b\u304b\u3051\u3066\u3001\u4e2d\u56fd\u6df1\u30bb\u30f3\u306b\u3066\u958b\u50ac\u3055\u308c\u307e\u3057\u305f2018JPNSEC International Workshop on Evolutionary Computation\u306b\u53c2\u52a0\u3044\u305f\u3057\u307e\u3057\u305f\u3002\u4eca\u56de\u306e\u30ef\u30fc\u30af\u30b7\u30e7\u30c3\u30d7\u306f\u3001\u9032\u5316\u8a08\u7b97\u5b66\u4f1a\u306b\u3088\u3063\u3066\u4e3b\u50ac\u3055\u308c\u305f\u56fd\u969b\u5b66\u4f1a\u3067\u3001\u9032\u5316\u8a08\u7b97\u5b66\u4f1a\u7814\u7a76\u4f1a\u3068\u3057\u3066\u306f14\u56de\u76ee\u3067\u521d\u306e\u6d77\u5916\u3067\u306e\u958b\u50ac\u3068\u306a\u308a\u307e\u3057\u305f\u3002\u4eca\u56de\u306f\u300129\u4ef6\u306e\u30dd\u30b9\u30bf\u30fc\u767a\u8868\u304c\u3042\u308a\u3001M1\u85e4\u4e95\u304c\u767a\u8868\u3057\u307e\u3057\u305f\u3002\u79c1\u306f1\u65e5\u306e15:20~17:20\u30672\u5206\u9593\u306e\u30d5\u30e9\u30c3\u30b7\u30e5\u30c8\u30fc\u30af\u53ca\u3073\u30dd\u30b9\u30bf\u30fc\u306b\u3088\u308b\u7814\u7a76\u767a\u8868\u3092\u884c\u3044\u307e\u3057\u305f\uff0e\u8133\u6a5f\u80fd\u3092\u9032\u5316\u8a08\u7b97\u3067\u89e3\u660e\u3059\u308b\u6211\u3005\u306e\u7814\u7a76\u767a\u8868\u306b\u306f\u3001\u591a\u304f\u306e\u65b9\u306b\u8db3\u3092\u904b\u3093\u3067\u9802\u304d\u3001\u69d8\u3005\u306a\u8b70\u8ad6\u3092\u4ea4\u308f\u3059\u3053\u3068\u304c\u3067\u304d\u307e\u3057\u305f\u3002
\n\u9032\u5316\u8a08\u7b97\u5b66\u4f1a\u30db\u30fc\u30e0\u30da\u30fc\u30b8\uff1a http:\/\/www.jpnsec.org\/symposium201802.html<\/p>\n- \n
- \u7814\u7a76\u767a\u8868\n
- \n
- \u767a\u8868\u6982\u8981<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n
\u79c1\u306f\u30011\u65e5\u306e15:20~17:20\u306b\u304b\u3051\u3066\u306e\u30dd\u30b9\u30bf\u30fc\u767a\u8868\u306b\u53c2\u52a0\u81f4\u3057\u307e\u3057\u305f\u3002\u767a\u8868\u5f62\u5f0f\u306f1\u4eba2\u5206\u306e\u30d5\u30e9\u30c3\u30b7\u30e5\u30c8\u30fc\u30af\u30681\u6642\u9593\u306e\u30dd\u30b9\u30bf\u30fc\u767a\u8868\u5f62\u5f0f\u3067\u884c\u308f\u308c\u307e\u3057\u305f\u3002
\n\u4eca\u56de\u306f\u3001\u8133\u6a5f\u80fd\u306e\u89e3\u660e\u306b\u304a\u3044\u3066\u8133\u72b6\u614b\u306e\u8b58\u5225\u7387\u3068\u8133\u90e8\u4f4d\u6570\u306b\u30c8\u30ec\u30fc\u30c9\u30aa\u30d5\u95a2\u4fc2\u3092\u60f3\u5b9a\u3057\u3001\u591a\u76ee\u7684\u6700\u9069\u5316\u554f\u984c\u3068\u3057\u3066\u5b9a\u7fa9\u3057\u3001NSGA-II\u306b\u3088\u308a\u6700\u9069\u5316\u3092\u884c\u3063\u305f\u7d50\u679c\u3092\u767a\u8868\u3057\u307e\u3057\u305f\u3002\u307e\u305f\u7372\u5f97\u3055\u308c\u305f\u30d1\u30ec\u30fc\u30c8\u89e3\u306e\u53d6\u308a\u6271\u3044\u306b\u304a\u3044\u3066\u3001\u5c11\u306a\u3044\u8133\u90e8\u4f4d\u6570\u3067\u306f\u8133\u72b6\u614b\u3092\u8868\u73fe\u3059\u308b\u306b\u306f\u4e0d\u5341\u5206\u3067\u3042\u308a\u3001\u591a\u3059\u304e\u308b\u5834\u5408\u306f\u904e\u5b66\u7fd2\u3092\u5f15\u304d\u8d77\u3053\u3057\u3066\u3044\u308b\u3068\u8003\u3048\u3001\u3088\u308a\u8133\u72b6\u614b\u8868\u73fe\u306e\u305f\u3081\u306e\u30a8\u30c3\u30bb\u30f3\u30b7\u30e3\u30eb\u306a\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3068\u3057\u3066\u3001\u69cb\u9020\u304c\u5171\u901a\u3057\u305f\u30b5\u30d6\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u62bd\u51fa\u3092\u884c\u3044\u307e\u3057\u305f\u3002
\n1\u6642\u9593\u3068\u3044\u3046\u767a\u8868\u306e\u4e2d\u3067\u3001\u5148\u751f\u65b9\u3084\u4ed6\u5927\u5b66\u306e\u5b66\u751f\u306b\u8cb4\u91cd\u306a\u610f\u898b\u3092\u9802\u304d\u307e\u3057\u305f\uff0e\u4ee5\u4e0b\u306b\u6284\u9332\u3092\u8a18\u8f09\u81f4\u3057\u307e\u3059\uff0e<\/p>\n\n\n
\n Our group conducted a research related to computational intelligence vigorously for a better understanding of brain functions. In the human brain, several brain regions are similarly activated and, accordingly, the task-process is executed. By using functional brain imaging techniques such as functional magnetic resonance imaging(fMRI), the activation of brain regions is measured, the degree of similarity of the obtained time series data is calculated, and brain regions that show similar activations are identified. Subsequently, the obtained information is represented by networks consisting of nodes, that show a similar activation and edges whose weights increased when the degree of similarity was is high.
\nThis network is called the functional brain network. The network representing brain states differ depending on the stimulus of each task and the network representation describes the corresponding brain states. In the conventional approaches for the finding of this kind of network, first, the region of interest (ROI) is set up as hypothesis and then the hypothesis is verified. Therefore, sufficient knowledge and experience are necessary for setting hypotheses. In our group, on the other hand, the network that can be one of the ROIs is extracted from the candidate network combination by using a data-driven approach. This extracted network is specially called an important functional brain network.
\nIn our data-driven approach, the functional brain imaging data of two states are prepared, and the important network that classify these two states with the highest accuracy is determined by numerical calculation. Thus, even if knowledge and experience are not sufficient, ROIs and hypotheses are set. However, in experiments for obtaining functional imaging data, in most cases, the number of obtained samples is smaller than the number of brain regions in the candidate network. In such cases, for the problem of classifying two states, there is a possibility that the network composed of a large number of brain regions is a network leading to oversampling. On the other hand, the accuracy of two states by classifying a network composed of a few brain regions is low. To solve this problem, the problem is formulated as a multi-objective optimization problem, whereby both the classification error and the number of brain regions are to be minimized, and the Pareto solution set is obtained by using an evolutionary computation method. The obtained Pareto solution set is composed of multiple networks. Sub-networks are explored, and a network composed of an appropriate number of brain regions is determined. By using this procedure, an important functional brain network for classifying two states is obtained.
\nIn this presentation, the outline of the problem, how the Pareto solution set is obtained, and how to decide the important functional brain network are explained for the resting- and meditative-state fMRI data. Furthermore, an initial analysis of the obtained functional brain network is performed.
\n <\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/p>\n
- \n
- \u8cea\u7591\u5fdc\u7b54<\/li>\n<\/ul>\n
\u4eca\u56de\u306e\u8b1b\u6f14\u767a\u8868\u3067\u306f\uff0c\u4ee5\u4e0b\u306e\u3088\u3046\u306a\u8cea\u7591\u3092\u53d7\u3051\u307e\u3057\u305f\uff0e
\n\u30fb\u8cea\u554f\u5185\u5bb9<\/strong>1<\/strong>
\n\u4eca\u56de\u8b58\u5225\u3057\u305f\u7791\u60f3\u6642\u3068\u5b89\u9759\u6642\u3068\u3044\u3046\u72b6\u614b\u9593\u306b\u306f\u305d\u3082\u305d\u3082\u5dee\u304c\u3042\u308b\u306e\u304b\u3001\u307e\u305f\u3069\u306e\u3088\u3046\u306a\u30bf\u30b9\u30af\u6642\u3067\u3082\u7cfb\u7d71\u56f3\u3092\u4f5c\u6210\u3067\u304d\u308b\u3088\u3046\u306a\u3001\u69cb\u9020\u304c\u5171\u901a\u3057\u305f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u5b58\u5728\u3059\u308b\u306e\u304b\u3068\u3044\u3046\u8cea\u554f\u3092\u53d7\u3051\u307e\u3057\u305f\u3002\u524d\u8005\u306b\u5bfe\u3057\u3066\u306f\uff0c\u7791\u60f3\u6642\u306b\u306f\u6ce8\u610f\u5236\u5fa1\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u95a2\u9023\u3057\u3001\u5b89\u9759\u6642\u306b\u306fDMN\u306a\u3069\u306e\u5b89\u9759\u6642\u7279\u6709\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u95a2\u9023\u3057\u3066\u304a\u308a\uff0c\u62bd\u51fa\u3055\u308c\u305f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u4e0a\u8a18\u306e\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u95a2\u9023\u3057\u305f\u9818\u57df\u304c\u542b\u307e\u308c\u3066\u3044\u308b\u3053\u3068\u304c\u671b\u307e\u3057\u3044\u3068\u7b54\u3048\u307e\u3057\u305f\u3002\u5f8c\u8005\u306b\u5bfe\u3057\u3066\u306f\u3001\u30d5\u30e5\u30fc\u30c1\u30e3\u30fc\u30ef\u30fc\u30af\u3067\u3042\u308b\u3068\u3044\u3046\u3053\u3068\u3001\u4e00\u65b9\u3067\u63d0\u6848\u624b\u6cd5\u306f\u3069\u306e\u3088\u3046\u306a\u8133\u72b6\u614b\u306b\u304a\u3044\u3066\u3082\u9069\u7528\u53ef\u80fd\u3068\u3044\u3046\u3053\u3068\u3092\u4f1d\u3048\u307e\u3057\u305f\u3002
\n
\n\u30fb\u8cea\u554f\u5185\u5bb9<\/strong>2<\/strong>
\n\u8133\u6a5f\u80fd\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u306f\u3001\u5404\u9818\u57df\u9593\u306e\u7269\u7406\u7684\u306a\u8ddd\u96e2\u306f\u5f71\u97ff\u3059\u308b\u306e\u304b\u3068\u3044\u3046\u8cea\u554f\u3092\u53d7\u3051\u307e\u3057\u305f\u3002\u8133\u6a5f\u80fd\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u89e3\u6790\u306b\u304a\u3044\u3066\u3001\u5404\u9818\u57df\u306e\u8133\u8840\u6d41\u5909\u5316\u91cf\u306b\u306f\u7269\u7406\u7684\u306a\u8ddd\u96e2\u304c\u5f71\u97ff\u3057\u306a\u3044\u3088\u3046\u306b\u51e6\u7406\u304c\u884c\u308f\u308c\u308b\u3068\u3044\u3046\u3053\u3068\u3001\u307e\u305f\u8133\u5185\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306b\u306f\u69cb\u9020\u7684\u306a\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u3084\u6a5f\u80fd\u7684\u306a\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u304c\u5b58\u5728\u3059\u308b\u3053\u3068\u3092\u8aac\u660e\u3057\u307e\u3057\u305f\u3002
\n <\/p>\n- \n
- \u611f\u60f3<\/li>\n<\/ul>\n
\u521d\u3081\u3066\u306e\u6d77\u5916\u3067\u306e\u767a\u8868\u3067\u3001\u4e8c\u5206\u9593\u306e\u30d5\u30e9\u30c3\u30b7\u30e5\u30c8\u30fc\u30af\u306f\u82f1\u8a9e\u3067\u767a\u8868\u3092\u884c\u3044\u307e\u3057\u305f\u3002\u65e5\u672c\u8a9e\u3067\u767a\u8868\u3059\u308b\u4e88\u5b9a\u3067\u6e96\u5099\u3057\u3066\u3044\u305f\u305f\u3081\u3001\u6025\u907d\u82f1\u8a9e\u306e\u539f\u7a3f\u3092\u6e96\u5099\u3059\u308b\u3053\u3068\u306b\u306a\u308a\u307e\u3057\u305f\u304c\u3001\u306a\u3093\u3068\u304b\u8a70\u307e\u3089\u305a\u767a\u8868\u3092\u7d42\u3048\u308b\u3053\u3068\u304c\u51fa\u6765\u307e\u3057\u305f\u3002\u666e\u6bb5\u304b\u3089GL\u73edMTG\u306e\u6e96\u5099\u3092\u82f1\u8a9e\u3067\u884c\u3063\u3066\u3044\u305f\u306e\u304c\u52a9\u3051\u306b\u306a\u3063\u305f\u306e\u3060\u3068\u601d\u3044\u307e\u3059\u3002\u4ed6\u306e\u767a\u8868\u8005\u306e\u30dd\u30b9\u30bf\u30fc\u767a\u8868\u3092\u8074\u8b1b\u3059\u308b\u4e2d\u3067\u3001\u81ea\u5206\u306b\u306f\u4e0d\u8db3\u3057\u3066\u3044\u308b\u77e5\u8b58\u304c\u591a\u3044\u3053\u3068\u3092\u518d\u78ba\u8a8d\u3059\u308b\u3053\u3068\u304c\u51fa\u6765\u307e\u3057\u305f\u3002\u3088\u304f\u8033\u306b\u3057\u305f\u306e\u306f\u3084\u306f\u308aDeep Learning\u3067\u3001\u8fd1\u5e74\u306e\u8a08\u7b97\u79d1\u5b66\u306b\u304a\u3044\u3066\u30db\u30c3\u30c8\u306a\u3053\u306e\u5206\u91ce\u306e\u77e5\u8b58\u306f\u6b20\u304b\u305b\u306a\u3044\u306a\u3068\u611f\u3058\u307e\u3057\u305f\u3002\u307e\u305f\u3001\u6700\u9069\u5316\u306e\u5404\u624b\u6cd5\u306e\u57fa\u672c\u7684\u306a\u77e5\u8b58\u306f\u8eab\u306b\u7740\u3051\u3066\u304a\u3053\u3046\u3068\u601d\u3044\u307e\u3057\u305f\u3002\u81ea\u5206\u304c\u3053\u308c\u304b\u3089\u5b66\u3076\u3079\u304d\u3053\u3068\u304c\u5206\u304b\u3063\u305f\u306e\u306f\u3001\u81ea\u5206\u306b\u3068\u3063\u3066\u3068\u3066\u3082\u5927\u304d\u306a\u6210\u9577\u3060\u3068\u601d\u3044\u307e\u3059\u3002\u305d\u3057\u3066\u5b66\u4f1a\u4e2d\u306b\u4ed6\u306e\u5927\u5b66\u306e\u5b66\u751f\u3068\u3082\u89aa\u3057\u304f\u306a\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3057\u305f\u3002\u6b21\u56de\u4ee5\u964d\u306e\u5b66\u4f1a\u3067\u306f\u6210\u9577\u3057\u305f\u59ff\u3067\u5148\u751f\u65b9\u3084\u5b66\u751f\u305f\u3061\u3068\u4f1a\u3048\u308b\u3088\u3046\u52aa\u529b\u3057\u3066\u3044\u304d\u305f\u3044\u3068\u601d\u3044\u307e\u3059\u3002<\/p>\n
- \n
- \u8074\u8b1b<\/li>\n<\/ol>\n
\u4eca\u56de\u306e\u5b66\u4f1a\u3067\u306f\uff0c\u4e0b\u8a18\u306e2\u4ef6\u306e\u767a\u8868\u3092\u8074\u8b1b\u3057\u307e\u3057\u305f\uff0e<\/p>\n
\n\n
\n \u767a\u8868\u30bf\u30a4\u30c8\u30eb\uff1aVegetation Evolution for Numerical Optimization
\n\u8457\u8005\uff1a Jun Yu and Hideyuki Takagi
\n\u30bb\u30c3\u30b7\u30e7\u30f3\u540d\uff1a Poster Presentations 2<\/strong>
\nThe main objective of this paper is to summarize a general mechanism from vegetation growth and reproduction, then develop a new population-based
\nevolutionary algorithm by simulating two different periods of vegetation, i.e. growth period and maturity period, to emphasize different optimization
\ncapabilities. The secondary one is to analyze the performance of the proposed algorithm as well as its applicability. Finally, we introduce some topics for open discussions.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n\u53bb\u5e74\u306e\u9032\u5316\u8a08\u7b97\u30b7\u30f3\u30dd\u30b8\u30a6\u30e0\u3067\u4e00\u5ea6\u304a\u8a71\u3057\u3057\u305f\u767a\u8868\u8005\u3067\u3042\u3063\u305f\u3053\u3068\u3001\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u304c\u690d\u7269\u306e\u6210\u9577\u3092\u6a21\u64ec\u3057\u3066\u3044\u308b\u3068\u3044\u3046\u3053\u3068\u3067\u3001\u3069\u306e\u3088\u3046\u306a\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306a\u306e\u304b\u8208\u5473\u304c\u3042\u3063\u305f\u305f\u3081\u8074\u8b1b\u3057\u307e\u3057\u305f\u3002\u79c1\u305f\u3061\u306e\u7814\u7a76\u5ba4\u3067\u3082\u5148\u6708\u306e\u6708\u4f8b\u767a\u8868\u4f1a\u3067\u3001\u751f\u7269\u9032\u5316\u306b\u30a4\u30f3\u30b9\u30d1\u30a4\u30a2\u3055\u308c\u305f\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u306b\u3064\u3044\u3066\u306e\u767a\u8868\u304c\u3042\u308a\u3001\u591a\u5c11\u306e\u4e8b\u524d\u77e5\u8b58\u304c\u3042\u3063\u305f\u305f\u3081\u697d\u3057\u304f\u8074\u8b1b\u3059\u308b\u3053\u3068\u304c\u3067\u304d\u307e\u3057\u305f\u3002\u63d0\u6848\u3055\u308c\u305f\u30a2\u30eb\u30b4\u30ea\u30ba\u30e0\u3068\u6bd4\u8f03\u3055\u308c\u3066\u3044\u305fDE\u3084PSO\u306b\u3064\u3044\u3066\u3001\u79c1\u81ea\u8eab\u7406\u89e3\u3092\u6df1\u3081\u3088\u3046\u3068\u601d\u3063\u305f\u826f\u3044\u304d\u3063\u304b\u3051\u306b\u3082\u306a\u308a\u307e\u3057\u305f\u3002
\n <\/p>\n\n\n
\n \u767a\u8868\u30bf\u30a4\u30c8\u30eb\uff1aPreliminary Study of Multi-Objective Air Traffic Optimization by using Step Back Cellular Automaton
\n\u8457\u8005\uff1a Katsuhiro Sekine\uff0cTomoaki Tatsukawa\uff0cShinsuke Nagaoka\u3001Kozo Fujii
\n\u30bb\u30c3\u30b7\u30e7\u30f3\u540d\uff1aPoster Presentations 2<\/strong>
\n\u6982\u8981: The air traffic demand is rapidly growing in re-cent years on the background of economic growth of Asia-Pacific area and popularization of Low-
\nCost Carriers. Every aircraft is planned to y along the flight plan. However, most of all aircrafts are not able to follow the scheduled plans due to various factors. One of the severe factors is the over-capacity of the arrival airport. When the number of aircraft exceeds the capacity of the airport, the air traffic controller does not allow air-planes to land. In this case, aircrafts have to take a detour, so-called “vectoring”, or keep making circles on the particular points, so-called “holding” to adjust the arrival time. The time adjustment such as vectoring and holding affects not only the delay of arrival time but also an increase in fuel consumption. If there is a efficient way to prepare a schedule with less time adjustment beforehand, it will be useful for future air traffic control. The objective in this study is to make efficient flight schedules with less congestion and enough resilience against traffic problems. In this study, we conduct a preliminary multi-objective optimization for air traffic ow by using NSGA-II. The number of objectives is three, and the number of
\ndesign variables is 579. For simulating air traffic ow, we adopt cellular-automaton-based model because air traffic ow not only consists of many complicated rules, but also strongly affected by installed equipment and weather.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n
\n\u81ea\u5206\u3068\u540c\u3058\u5b9f\u554f\u984c\u3078\u306e\u9032\u5316\u7684\u8a08\u7b97\u306e\u9069\u7528\u3092\u884c\u3063\u305f\u7814\u7a76\u306e\u305f\u3081\u8074\u8b1b\u3057\u307e\u3057\u305f\u3002\u76ee\u7684\u95a2\u6570\u3068\u3057\u3066\u3001\u98db\u884c\u6a5f\u306e\u5e73\u5747\u9045\u5ef6\u6642\u9593\u306e\u6700\u5c0f\u5316\u3001\u9045\u5ef6\u3059\u308b\u98db\u884c\u6a5f\u306e\u6bd4\u7387\u306e\u6700\u5c0f\u5316\u3001\u5e73\u5747\u71c3\u6599\u6d88\u8cbb\u91cf\u306e\u6700\u5c0f\u5316\u30923\u76ee\u7684\u306b\u8a2d\u5b9a\u3057\u3001\u81ea\u5206\u3068\u540c\u3058NSGA-\u2161\u3067\u6700\u9069\u5316\u3092\u884c\u3063\u3066\u3044\u307e\u3057\u305f\u3002\u6700\u9069\u5316\u306e\u7d50\u679c\u3068\u3057\u3066\u7d043\u5206\u9593\u306e\u6642\u9593\u306e\u524a\u6e1b\u306b\u6210\u529f\u3057\u305f\u3068\u5831\u544a\u3055\u308c\u3066\u3044\u307e\u3057\u305f\u3002\u307e\u3060\u7814\u7a76\u304c\u4e88\u5099\u5b9f\u9a13\u306e\u6bb5\u968e\u3067\u3001\u4eca\u5f8c\u306e\u7814\u7a76\u3067\u306f\u7372\u5f97\u3055\u308c\u305f\u30d1\u30ec\u30fc\u30c8\u89e3\u306e\u53d6\u308a\u6271\u3044\u65b9\u3084\u3001\u60aa\u5929\u5019\u306b\u3088\u308b\u9045\u5ef6\u3001\u30a8\u30f3\u30b8\u30f3\u30c8\u30e9\u30d6\u30eb\u306b\u3088\u308b\u9045\u5ef6\u306a\u3069\u3082\u8003\u616e\u3057\u3066\u3086\u304f\u3068\u304a\u3063\u3057\u3083\u3063\u3066\u3044\u307e\u3057\u305f\u3002\u767a\u8868\u8005\u304cB4\u306e\u5b66\u751f\u3060\u3063\u305f\u305f\u3081\u4eca\u5f8c\u306e\u5b66\u4f1a\u3067\u3069\u306e\u3088\u3046\u306a\u9032\u6357\u304c\u3042\u3063\u305f\u306e\u304b\u3092\u805e\u304f\u3053\u3068\u304c\u697d\u3057\u307f\u3067\u3059\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"2018\u5e748\u670831\u65e5~9\u67081\u65e5\u306b\u304b\u3051\u3066\u4e2d\u56fd\u6df1\u30bb\u30f3\u5357\u65b9\u79d1\u6280\u5927\u5b66\u306b\u3066\u958b\u50ac\u3055\u308c\u307e\u3057\u305f2018JPNSEC International Workshop on Evolutionary Computation\u306b\u53c2\u52a0\u3044\u305f\u3057\u307e\u3057\u305f\uff0e … <\/p>\n
- \u8074\u8b1b<\/li>\n<\/ol>\n
- \u611f\u60f3<\/li>\n<\/ul>\n
- \u8cea\u7591\u5fdc\u7b54<\/li>\n<\/ul>\n
- \u767a\u8868\u6982\u8981<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n
- \u7814\u7a76\u767a\u8868\n
- \u8b1b\u6f14\u4f1a\u306e\u8a73\u7d30<\/li>\n<\/ol>\n
![\"\"](\"http:\/\/www.is.doshisha.ac.jp\/news\/wp-content\/uploads\/2018\/08\/0119_xlarge.jpg\")
<\/a> ![\"\"](\"http:\/\/www.is.doshisha.ac.jp\/news\/wp-content\/uploads\/2018\/08\/0121_xlarge.jpg\")
<\/a>