problem n. 1.問題,課題;疑難問題;令人困惑的情況。 2.【數(shù)、物】習(xí)題;作圖題。 3.(象棋的)布局問題。 the problem of unemployment 失業(yè)問題。 His whole conduct is a problem to me. 他的一切行為我都不理解。 adj. 1.成問題的;難處理的。 2.關(guān)于社會(huì)問題的。 a problem child 【心理學(xué)】問題兒童;難管教的孩子。 a problem novel [play] (反映社會(huì)問題等的)問題小說[戲劇]。 sleep on [upon, over] a problem 把問題留到第二天解決。
minimum n. (pl. minimums, -ma ) 最小,最低,最少限度;【數(shù)學(xué)】極小(值)。 the irreducible minimum 無法減少的最小限度。 The thermometer reached the minimum for the year. 寒暑表降到當(dāng)年最低度數(shù)。 adj. 最少的,最小的,最低的。 the minimum value 【數(shù)學(xué)】極小值。
We proposed an improved simulated annealing algorithm with neighbor function based on self - optimization of scale parameter . furthermore incorporating disaster - modification and the improved annealing into genetic algorithm , an improved genetic - annealing algorithm is proposed . in order to solve the deceptive minimum problem , an improved evolutionary strategy combined with similarity detection and improved mutation operator 提出了鄰域尺度函數(shù)自尋優(yōu)的模擬褪火算法,結(jié)合遺傳算法,引入災(zāi)變算子,提出了改進(jìn)遺傳模擬褪火算法;為了解決尋優(yōu)過程中的最小欺騙問題,我們提出了相似性檢測(cè),結(jié)合改進(jìn)的適應(yīng)值無關(guān)變異算子,提出了基于相似性檢測(cè)和適應(yīng)值無關(guān)變異算子的進(jìn)化策略算法。
2 . on the base of detailedly analysing the fourier neural networks , we find this neural networks have the characteristic which can transform the nonlinear mapping into linear mapping . so , we improve the original learning algorithm based on nonlinear optimization and propose a novel learning algorithm based on linear optimization ( this dissertation adopts the least squares method ) . the novel learning algorithm highly improve convergence speed and avoid local minima problem . because of adopting the least squares method , when the training output samples were affected by white noise , this algorithm have good denoising function 在詳細(xì)分析已有的傅立葉神經(jīng)網(wǎng)絡(luò)的基礎(chǔ)上,發(fā)現(xiàn)傅立葉神經(jīng)網(wǎng)絡(luò)具有將非線性映射轉(zhuǎn)化成線性映射的特點(diǎn),基于這個(gè)特點(diǎn),對(duì)該神經(jīng)網(wǎng)絡(luò)原有的基于非線性優(yōu)化的學(xué)習(xí)算法進(jìn)行了改進(jìn),提出了基于線性優(yōu)化方法(本文采用最小二乘法)的學(xué)習(xí)算法,大大提高了神經(jīng)網(wǎng)絡(luò)的收斂速度并避免了局部極小問題;由于采用了最小二乘方法,當(dāng)用來訓(xùn)練傅立葉神經(jīng)網(wǎng)絡(luò)的訓(xùn)練輸出樣本受白噪聲影響時(shí),本學(xué)習(xí)算法具有良好的降低噪聲影響的功能。