收斂速率 的英文怎麼說

中文拼音 [shōuliǎnsùlǜ]
收斂速率 英文
convergence rate
  • 收 : Ⅰ動詞1 （把攤開的或分散的事物聚集、合攏） put away; take in 2 （收取） collect 3 （收割） harvest...
  • 斂 : Ⅰ動詞1 （收起; 收住） hold back; keep back 2 （約束） restrain 3 （收集; 徵收） gather; collect; ...
  • 速 : Ⅰ形容詞（迅速; 快） fast; rapid; quick; speedy Ⅱ名詞1 （速度） speed; velocity 2 （姓氏） a surna...
  • 率 : 率名詞（比值） rate; ratio; proportion
  • 收斂 : 1 （減弱或消失） weaken or disappear 2 （約束言行） restrain oneself 3 [數學] convergence; constr...
  • 速率 : speed; rate; tempo
  1. The method proposed in this thesis do well in solving the problems of multi - damping - ratio - spectra simulation. it is convenient to obtain the pareto optimal solution set of the multi - object question by using implicit parallel genetic algorithms and the method can meet the practical needs for simulating ground motions coinciding with multi - damping - ratio - spectra in seismic design. the crossing rate and variance rate are important parameters of genetic algorithms which affect the rate of convergence, the adapting rate of cross and variation in this paper can auto - adapt and according to stand or fall of current sample, it assures the sample approach to the pareto optimal solution set in fast convergent speed

    較好地解決多阻尼比反應譜擬合問題；本文方法通過一次運行就能獲得一組具有集系特性的地震動，在擬合多阻尼比反應譜的人造地震波集系的模擬方面有傳統方法所不能比擬的優勢，產生的人造波或人造波集系可滿足工程抗震設計需要；在遺傳演算法中，交叉概率和變異概率是影響收斂速度的重要參數，本文採用的改進自適應交叉概率和變異概率，可以根據當前樣本的好壞程度來自動地選擇適當的交叉概率和變異概率，以保證演算法始終以較好的速度向pareto最優解集逼近。
  2. Based on fractional sampling method, a new super - exponential iteration decision feedback blind equalization algorithm for severely nonlinear phase distortion channels was proposed

    因此針對嚴重頻率衰落和非線性相位失真通道，提出了一種分數采樣的混合盲均衡演算法，並獲得了較快的收斂速度和較小的剩餘均方誤差。
  3. Finally, bp neural network is improved for face recognition, the problem on choice of parameters is discussed, the sigmoid function and weight adjustment are improved for higher convergence speed

    討論了傳統bp神經網路的參數選取問題，對sigmoid函數和網路學習速率進行了改進，以提高系統的收斂速度和收斂率。
  4. In this dissertation, we firstly prove that any dirichlet problem is indeed equal to a voltages problem of networks. we give five solutions to dirichlet problem in two dimensions ; among these five solutions, we prove that the iteration solution and the solution of relaxations are exponential convergence, then we estimate their respective convergence rates ; secondly, we discuss random walks on general networks, prove that there is an one to one correspondence between networks and reversible ergodic markov chains ; thirdly, we give probabilistic interpretation of voltages for general networks : when a unit voltage is applied between a and b, making va = 1 and vb = 0, the voltage vx at any point x represents the probability that a walker starting from x will return to a before reaching b ; furthermore, we study the relationship between effective resistance and escape probability : starting at a, the probability that the walk reaches b before returning to a is the ratio of the effective conductance and the total conductance

    本文證明了任何邊值的dirichlet問題都可轉化為求解電路電壓的問題：給出了計算平面格點上dirichlet問題的5種方法：證明了迭代法和松馳法都是指數收斂的，並分別給出收斂速度的估計；討論了一般電路上的隨機徘徊，驗證了電路與可逆的遍歷markov鏈是一一對應的；給出了電路電壓的概率解釋：當把1伏電壓加於a ， b兩端，使得v _ a = 1 ， v _ b = 0時，則x點的電壓v _ x表示對應的markov鏈中，從x出發，到達b之前到達a的概率；進一步地，給出了逃離概率與有效電阻之間的關系：從a出發，在到達b之前到達a的概率為有效傳導率與通過a的總傳導率之比。
  5. The results of numerical experiment are reported to show the effectiveness of the proposed algorithms. reduced hessian algorithm has became one of the very popular and most effecive methods for solving nonlinear equality constrained programming. recently, chaya gur - witz proposed the two - piece update of a projected hessian matrix

    最近， chayagurwitz在nocedal和overton等人工作的基礎上，提出了兩塊校正的投影hesse矩陣方法，討論了其演算法的局部收斂速率，但未涉及整體收斂性。
  6. Compared with the classical bp algorithm, robust adaptive bp algorithm possesses some advantages as following : ( 1 ) increasing the accuracy of the network training by means of using both the relative and absolute residual to adjust the weight values ; ( 2 ) improve the robustness and the network convergence rate through combining with the robust statistic technique by way of judging the values of the samples " relative residual to establish the energy function so that can suppress the effect on network training because of the samples with high noise disturbances ; ( 3 ) prevent entrapping into the local minima area and obtain the global optimal result owing to setting the learning rate to be the function of the errors and the error gradients when network is trained. the learning rate of the weights update change with the error values of the network adaptively so that can easily get rid of the disadvantage of the classical bp algorithm that is liable to entrap into the local minima areas

    與基本bp演算法相比，本文提出的魯棒自適應bp演算法具有以下優點： （ 1 ）與魯棒統計技術相結合，通過訓練樣本相對偏差的大小，確定不同訓練樣本對能量函數的貢獻，來抑制含高噪聲干擾樣本對網路訓練的不良影響，從而增強訓練的魯棒性，提高網路訓練的收斂速度； （ 2 ）採用相對偏差和絕對偏差兩種偏差形式對權值進行調整，提高了網路的訓練精度； （ 3 ）在採用梯度下降演算法對權值進行調整的基礎上，通過將學習速率設為訓練誤差及誤差梯度的特殊函數，使學習速率依賴于網路訓練時誤差瞬時的變化而自適應的改變，從而可以克服基本bp演算法容易陷入局部極小區域的弊端，使訓練過程能夠很快的「跳出」局部極小區域而達到全局最優。
  7. Abstract : based on the iterative bit - filling procedure, a computationally efficient bit and power allocation algorithm is presented. the algorithm improves the conventional bit - filling algorithms by maintaining only a subset of subcarriers for computation in each iteration, which reduces the complexity without any performance degradation. moreover, a modified algorithm with even lower complexity is developed, and equal power allocation is introduced as an initial allocation to accelerate its convergence. simulation results show that the modified algorithm achieves a considerable complexity reduction while causing only a minor drop in performance

    文摘:基於迭代的比特和功率分配機制，提出了一種低復雜度的比特和功率分配演算法.與傳統的迭代分配演算法不同，該演算法在每次迭代中只需要比較幾個特定的子載波.該方法在保持傳統迭代演算法性能的前提下極大地減小了迭代分配演算法的復雜度.此外，通過選擇等功率分配方案作為初始方案加快了演算法的收斂速度，進一步降低了演算法復雜度.模擬結果表明，提出的改進演算法在基本不犧牲系統性能的前提下有效地降低了演算法復雜度
  8. Because ga possesses the traits of can global random search, the robustness is strong, been use briefly and broadly, it didn ’ t use path search, and use probability search, didn ’ t care inherence rule of problem itself, can search the global optimum points effectively and rapidly in great vector space of complicated, many peak values, cannot differentiable. so it can offset the shortages of nn study algorithm, can reduce the possibility that the minimum value get into local greatly, the speed of convergence can improve, interpolation time shorten greatly, the quantity of training reduce

    因為遺傳演算法具有全局隨機搜索能力，魯棒性強、使用簡單和廣泛的特點，它不採用路徑搜索，而採用概率搜索，不用關心問題本身的內在規律，能夠在復雜的、多峰值的、不可微的大矢量空間中迅速有效地尋找到全局最優解，所以可以彌補神經網路學習演算法的不足，使陷入局部最小值的可能性大大減少，使得收斂速度提高，訓練量減小。
  9. Simple genetic algorithm gets local minimization too easily and converges slowly. to solve these problems, adaptive crossover rate that has reverse hyperbolic rel ation with the numbers of iteration is designed, and adaptive mutation rate that has reverse proportion to the distances of parents and reverse exponential relat ion to the numbers of iteration is put forward. the practical simulation results show that the adaptive ga has greater convergence speed and larger probability o f getting the best solution

    簡單遺傳演算法存在著收斂速度慢、易陷入局部極小等缺陷.針對這些缺陷，本文設計出隨相對遺傳代數呈雙曲線下降的自適應交換率，並提出與父串間的相對歐氏距離成反比、隨相對遺傳代數指數下降的自適應變異率.實例驗證表明，具有自適應交換率和變異率的遺傳演算法在收斂速度和獲得全局最優解的概率兩個方面都有很大的提高
  10. The artificial neural net ( ann ) way is universal regard as one of the most effective ways of stlf. in this paper, some research is developed for stlf using ann ways in several parts : the first part is about the arithmetic of ann based on bp model, namely the advanced of traditional bp arithmetic, one alterable step and scale bp arithmetic based on comparability of model and probability of accepting bp arithmetic is used to enhances a lot the convergence rate of learning process of bp network, but also avoid the stagnation problem to some extent. it indicates that the ann ' s efficiency and precision by the way can be ameliorated by the simulation of real data

    神經網路方法在短期預測中已經被公認為較有效的方法，本文針對神經網路用於電力系統短期負荷預測的幾個方面展開研究工作：第一部分研究一般用於負荷預測的神經網路bp模型的演算法，即對傳統的bp演算法的改進，將一種基於模式逼近度和接受概率的變步長快速bp演算法應用到短期負荷預測，模擬結果表明該方法有效的改善了bp演算法收斂速度慢以及容易陷入局部最小點的缺點，從而提高了神經網路用於負荷預測的效率和精度。
  11. What flow is that, we use model simulation to analyze the em algorithm contraction ratio. through network simulating, we analyze the factors which can influence loss inference algorithm accuracy like measurement strategy or routing algorithm. we analyze the accuracy and contraction characteristic of multicast - based direct algorithm and em algorithm, and compare the error factor between them

    實驗中通過網路模擬模型，確定了em演算法的收斂速率;研究了不同測量策略和路由器擁塞避免演算法對丟包率推理演算法準確率的影響;分析了單點多播的de和em演算法準確性、收斂性等特徵，通過比較兩種演算法的統計誤差，得出em演算法略優于de演算法的結論。
  12. This model is simple and easy to be implemented. the refined genetic algorithm is also set up, in which some improvements are made on chromosome coding, fitness fuction, stopping rule, crossover and mutation pattern

    對遺傳演算法的染色體編碼、群體規模的確定、收斂準則、交叉、變異等環節進行了改進，提高了演算法的全局尋優概率和收斂速度。
  13. Based on the discussions of the conventional and recent methods of short term load forecasting such as time series, multiple regression approaches and artificial intelligence technologies, this paper presents a hybrid short term forecasting model which combines the artificial neural network ( ann ) and genetic algorithm ( ga ). in order to improve the convergence speed and precision of the back - propagation ( bp ), a new improved algorithm - the adapted learning algorithm based on quasi - newton method is given

    本文首先分析比較了電力系統短期負荷預測的傳統方法時間序列法和回歸方法以及最近的專家系統和神經網路技術的優點和不足，然後針對人工神經網路bp演算法的不足對其進行了改進，採用了基於擬牛頓的自適應演算法，它提高了網路學習效率，具有較快的收斂速度和較高的精度。接著提出了改進的遺傳演算法來改善神經網路的局部收斂性。
  14. In this paper, we give a kernel shape estimation of m ( x ) using variable bandwidth local linear refression approch, and discuss the asymptotic normality, the convergence rate of mean square and convergence rate with probability

    本文對上述模型，利用變窗寬局部線性回歸方法，給出了m （ x ）的核形估計，並討論了這一估計的漸近正態性、依概率收斂速度、和均方收斂速度。
  15. The main contributions of this dissertation are summarized as follow : ( 1 ) an ilc approach combining feedforward with current feedback is developed based on optimal feedback control and the gradient method. a sufficient condition that guarantees the convergences is given for linear system. the procedures of designing the algorithm can employ lqr, h2 or h approaches to improve the convergence rate of learning in iterations

    本文的主要成果有： 1 、在開閉環綜合迭代學習控制結構的基礎上，分析了利用梯度下降法設計前饋迭代學習控制器時，為保證演算法的收斂性，閉環控制系統應該滿足的充分條件，並依據提高演算法收斂速率的優化條件，給出了基於lqr 、 h _ 2和h等優化控制技術的迭代學習控制演算法的設計方法。
  16. By giving some properties of the double dogleg path, we prove the global convergence and fast local convergence rate of the proposed algorithm under some reasonable conditions

    通過給出雙折線路徑的一些性質，在合理的條件下，我們證明了所提供演算法具有整體收斂性並得到局部收斂速率。
  17. Specifically in the field of institutional change and economic growth, the model forecasts that imposed - change and induced - change not only has level effects and growth effects on economic growth, but also has significant impact on the convergent speed

    模型顯示，經濟體系不斷積累的制度存量是驅動增長的重要動因：不僅對增長具有水平效應和增長效應，而且對轉移進程的收斂速率產生重要影響。
  18. We show the proposed algorithm is globally convergen t and locally fast convergent rate even if conditions are reasonalbe

    在合理的假設條件下，我們證明了這一演算法不僅具備整體收斂性，而且具有超線性收斂速率。
  19. Proposes a new algorithm for estimating the vector channel and the optimum beamforming weights. studies its tracking performance under realistic channel conditions

    通過分析和天線方向圖模擬，證明新演算法在收斂速率和復雜度上有較大改善。
  20. Discussion and testifying were made to the convergence of the algorithm under the condition of having constrains in objects " outputs. one new algorithm, which can maintain the convergent speed under such constraint conditions, was presented. using iterative learning control in real job of industrial robotic manipulators, we h

    本文對對象輸出有限制情況下迭代學習演算法的收斂性做了討論和證明，並且提出了一種在這種情況下，能相對維持收斂速率的迭代學習律的改進策略。
分享友人
分享到 Line

分享到臉書