Search - im - CodeBus

Search - im - List

[AI-NN-PR] Bayesnet DL : 0: a Java toolkit for training, testing, and applying Bayesian Network Classifiers. Implemented classifiers have been shown to perform well in a variety of artificial intelligence, machine learning, and data mining applications. -a Java toolkit for training, testing, and applying Bayesian Network Classifiers. Im plemented classifiers have been shown to perfo rm well in a variety of artificial intelligence , machine learning, and data mining applications.
Date : 2025-12-24 Size : 492kb User : lyb
[AI-NN-PR] edrk DL : 0: 主要包括免疫识别、免疫学习、免疫记忆、克隆选择、个体多样性、分布式和自适应等,-It is the real engineering app licat ion s that draw the b road at ten2 t ion of compu ter scien t ist s to recogn ize the great po ten t ial of A IS, hereby som e impo rtan t app li2 cat ion f ields as info rm at ion secu rity, pat tern recogn it ion, op t im izat ion, m ach ine learn ing, data m in ing, robo t ics, diagno st ics and cybernet ics etc. are review ed
Date : 2025-12-24 Size : 1kb User : 小蓝
[AI-NN-PR] AHybidGeneticAgorithmtoSolveTSPandMTSP DL : 0: 求解TSP和MTSP的混合遗传算法_英文_-Abstract:M any app licat ions are invo lved w ith mult ip le salesmen each of w hom visits a subgroup cit ies and returns the same start ing city. The to tal length of all subtours is required to be m ini2 mum. Th is is calledM ult ip le T raveling Salesmen P roblem (M TSP). There are various heurist ic methods to obtain op t imal o r near2op t imal so lut ions fo r the TSP p roblem. But to the M ult ip le T raveling Salesmen P roblem , there are no t much app roaches to so lveM TSP. In th is paper, a hy2 brid genet ic algo rithm to so lve TSP and M TSP is p resented. Th is algo rithm combines GA and heurist ics. N umerical experiments show that the new algo rithm is very efficient and effect ive. Key words: TSP op t im izat ion genet ic algo rithm 2op t
Date : 2025-12-24 Size : 212kb User : Notics
[AI-NN-PR] Adaptive-Hysteresis DL : 0: 基于径向基函数神经网络迟滞非线性自适应控制提出了一种新的动态迟滞非线性模型. 将一定数量不同死区宽度的 backlash 模型并行相加, 作为一个动态系统以仿真执行器中的迟滞特性. 利用该模型, 采用伪控制方法设计了一套具有未知迟滞特性非线性系统的神经网络自适应控制方案, 通过自适应算法来调整干扰项的上限. 采用 Lyapunov 稳定性理论进行了严格证明, 仿真试验验证了所提方案的有效性.- A nov el class of hysteresis mo dels w as proposed. A cer tain num ber o f different deadband w idth backlash models are superposed, w hich represents a dynamics to m im ic hysteresis in the actuator. With the mo del proposed, an radial basis function neural netw ork ( RBFN )-based adaptive control scheme for nonlinear sy stems w ith unknow n hysteresis nonlinearity w as dev elo ped. The control scheme adopts the de- sign method of pseudo-co ntro l. Witho ut the assumption of boundedness of disturbance term , it is tuned thr oug h adaptive algo rithm . The stability is rigidly pr oved v ia Lyapunov theory and the effectiveness of the pro posed contr ol scheme is illustrated through simulatio n.
Date : 2025-12-24 Size : 207kb User :
[AI-NN-PR] adaptive-genetic-algorithm DL : 0: 自适应GA SVM 参数选择算法研究Param eter selection algorithm for support vector machines based on adaptive genetic algorithm 支持向量机是一种非常有前景的学习机器, 它的回归算法已经成功地用于解决非线性函数的逼近问题. 但是, SVM 参数的选择大多数是凭经验选取, 这种方法依赖于使用者的水平, 这样不仅不能获得最佳的函数逼近效果, 而且采用人工的方法选择 SVM 参数比较浪费时间, 这在很大程度上限制了它的应用. 为了能够自动地获得最佳的 SVM 参数, 提出了基于自适应遗传算法的 SVM 参数选取方法. 该方法根据适应度值自动调整交叉概率和变异概率, 减少了遗传算法的收敛时间并且提高了遗传算法的精度, 从而确保了 SVM 参数选择的准确性. 将该方法应用于船用锅炉汽包水位系统建模, 仿真结果表明由该方法所得的 SVM 具有较简单的结构和较好的泛化能力, 仿真精度高, 具有一定的理论推广意义.- The support vector m achine ( SVM ) is a prom ising artificia l inte lligence technique, in w hich the regres sion a lgorithm has a lready been used to so lve the nonlinear function approach successfully. Un fortunate ly, m ost us ers se lect param eters for an SVM by ru le o f thumb, so they frequently fail to generate the optim al approach ing e ffect for the function. Th is has restricted effective use o f SVM to a great degree. In order to get optim a l param eters auto m atically, a new approach based on an adaptive genetic a lgorithm ( AGA ) is presented, w hich autom atica lly ad justs the param eters for SVM. This m ethod se lects crossover probability and mutation probab ility accord ing to the fitness va lues of the object function, therefo re reduces the convergence tim e and im proves the prec ision o fGA, in suring the accuracy of param eter se lection. Th is m ethod w as applied to m odeling of w ater level system o f a sh ip b
Date : 2025-12-24 Size : 324kb User :
[AI-NN-PR] duoquanzhishengjingwangluo DL : 0: 应用多权值神经网络方法对静态手势进行识别, 对手势字母图像采用傅里叶描述子提取特征信息, 取低频信息成分构建成犯维特征向量, 并应用多权值神经网络的算法, 构建各类的神经元网络-W ith th e develo Pm en t of hu m an eom p uter intera etion te ehn olo盯, th e h as been b ased on an im P o rt a n t tas k fo r U r o n s diseu ssion . Th is Pap er pro Posed a new m eth o d fo r re searc h of un ders tan ding fo r hu m an 􀀁 5 gestu re re eogn izing of han d al Ph ab et gesture s, w hieh m ulti 一we ig hted ne al 即rith m of m ulti一w eigh ted n e U r o n S Th e 5 fo r m eth od ap plied Fo uri er一d eseri Ptors fo r fe atu re extra etion an d eolb ined wi th the elassifi eation re eognizin g , th e re sul ts sh ow ed th at th e m eth od p erfo rm ed w el .
Date : 2025-12-24 Size : 563kb User : 流星
[AI-NN-PR] Classification-based-- DL : 0: 多边多议题谈判最复杂的现实的协商问题。自动AP- 接近已被证明特别看好复杂的东北 gotiations和以前的研究表明进化COM 可能是有用putation的这种复杂的系统。要改善证明效率的多边多议题的现实NE- gotiations，避免完整信息的要求关于谈判的重刑模型的基础上，一种新的谈判提出了一个改进的进化算法P-ADE。-Abstract Multi-lateral multi-issue negotiations are the most complex realistic negotiation problems. Automated ap- proaches have proven particularly promising for complex ne- gotiations and previous research indicates evolutionary com- putation could be useful for such complex systems. To im- prove the eﬃ ciency of realistic multi-lateral multi-issue ne- gotiations and avoid the requirement of complete informa- tion about negotiators, a novel negotiation model based on an improved evolutionary algorithm p-ADE is proposed. The
Date : 2025-12-24 Size : 562kb User : 曲会晨