stochastic wavelet neural networks as approximators of a kind of stochastic processes 基于隨機(jī)小波神經(jīng)網(wǎng)絡(luò)的一類隨機(jī)過程的逼近
comparison and analysis of sufficient conditions for two classes fuzzy systems as universal approximators 兩類模糊系統(tǒng)作為通用逼近器的充分條件的比較與分析
it is pointed that universal approximating property of fuzzy controllers depends on the expressions or values of ( ?, 0 ) and ( ?, 1 ), and based on it, the sufficient conditions for fuzzy controllers being universal approximators are proposed 摘要通過分析模糊控制器的一般算法,發(fā)現(xiàn)模糊控制器是否具有泛逼近性,關(guān)鍵取決于模糊蘊(yùn)涵算子(?,0)和(?,1)時(shí)的表達(dá)式或取值。
in the research of the algorithms and theory of temporal difference learning, a new class of multi-step learning prediction algorithms based on linear function approximators and recursive least squares methods is proposed, which are called the rls-td ( t ) learning algorithm . the convergence with probability one of the rls-td ( t ) algorithm is proved for ergodic markov chains, and the conditions for convergence are analyzed 在時(shí)域差值學(xué)習(xí)(temporaldifferencelearning)學(xué)習(xí)算法和理論方面,首次提出了一種基于線性值函數(shù)逼近的多步遞推最小二乘td()(rls-td())學(xué)習(xí)算法,并分析和證明了該算法在求解遍歷markov鏈學(xué)習(xí)預(yù)測(cè)問題中的收斂條件和一致收斂性。
radial basis function ( rbf ) neural networks, which can be used as non-linear classifiers, have been applied extensively in pattern classification because of their salient features in being universal approximators, possing the best approximation property and the fast learning speed and having more compact topology than other neural networks 徑向基(radialbasisfunction,rbf)神經(jīng)網(wǎng)絡(luò)分類器,作為非線性分類器,具有全局函數(shù)逼近、函數(shù)擬合度好、收斂速度快、網(wǎng)絡(luò)結(jié)構(gòu)緊湊等優(yōu)點(diǎn),被廣泛運(yùn)用于模式識(shí)別問題中。