以多觀點進化演算法訓練TSK式模糊類神經網路之研究

Abstract

在這篇論文中，我們主要探討的是進化演算法的多觀點求解機制。進化演算法因具有全域最佳解搜尋的能力，近來已成為熱門的研究領域。如何有效提高演化的效率與增加得到最佳解的機會已成為此領域重要之議題，因此我們提出一個新的進化演算法，稱為多觀點進化演算法，以解決此上述的議題，並利用提出的多觀點進化演算法來求解模糊類神經網路之參數。不同於傳統的進化演算法，我們提出的多觀點進化演算法不僅可考慮個體部分的解以增加解的多樣性外，並可提供較合適的搜尋空間以期提高得到最佳解的機會。實驗結果顯示，我們提出的多觀點進化演算法可以有效地藉由最佳化模糊神經網路的參數找到輸入與輸出變數間之對映關係。此外，我們亦採用時間序列與真實世界的應用驗證所提出的方法，相較於其他的研究，多觀點進化演算法不僅有較佳的效能外，並具有較平滑的收斂曲線以得到較佳的解。這些結果也顯示，我們所提出的方法可以克服複雜的非線性問題，同時基於此演算法下，我們亦成功發展一個可偵測使用者學習迷惘問題的系統架構。

In this dissertation, we mainly focus on researching the multiple angle evolutionary algorithms. An evolutionary computation has become a popular research field due to its global searching ability. Therefore, it raises a challenge to develop an efficient and robustness evolutionary algorithm to not only reduce the evolution process but also increase the chances to meet the global solution. To this end, this study aims to provide a novel evolutionary algorithm named multiple angle evolutionary algorithm to address this issue; the proposed algorithm is applied to adjust the parameters of the neuro-fuzzy networks. The proposed evolutionary algorithm can consider the influence of partial solutions and provide a suitable searching space to increase the chances to meet the global solution. As shown in the results, the proposed evolutionary algorithm obtains better performance and smoother learning curves than other existing evolutionary algorithms. In other words, the proposed algorithm can efficiently adjust the parameters of the neuro-fuzzy networks to find the suitable relationship between the input patterns and outcomes by considering multiple angles. Base on the experimental results, a framework is proposed to build a benchmark for developing the evolutionary algorithms that cannot only consider the influence of partial solutions but also provide a suitable searching space. Moreover, it can also be successfully applied in both simulation and real world applications. It implies that the proposed algorithm is suitable for overcome the complex nonlinear problems. Also, based on the framework, we can also develop a novel architecture to automatically detect users’ disorientation problems.