Description: FASBIR(Filtered Attribute SUBSPACE based Bagging with Injected Randomness) is a variant of Bagging algorithm, whose purpose is to improve accuracy of local learners, such as kNN, through multi-model perturbing ensemble. Reference: Z.-H. Zhou and Y. Yu. Ensembling local learners through multimodal perturbation. IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics, 2005, vol.35, no.4, pp.725-735.
標(biāo)簽： Description Randomness Attribute Filtered
上傳時(shí)間： 2015-04-10
上傳用戶：ynzfm
A new blind adaptive multiuser detection scheme based on a hybrid of Kalman filter and SUBSPACE estimation is proposed. It is shown that the detector can be expressed as an anchored signal in the signal SUBSPACE and the coefficients can be estimated by the Kalman filter using only the signature waveform and the timing of the desired user.
標(biāo)簽： multiuser detection adaptive SUBSPACE
上傳時(shí)間： 2015-09-07
上傳用戶：xieguodong1234
％ COMPDIR Computes a search direction in a SUBSPACE defined by Z. ％ Helper function for NLCONST. ％ Returns Newton direction if possible. ％ Returns random direction if gradient is small. ％ Otherwise, returns steepest descent direction. ％ If the steepest descent direction is small it computes a negative ％ curvature direction based on the most negative eigenvalue. ％ For singular matrices, returns steepest descent even if small.
標(biāo)簽： Z. direction Computes function
上傳時(shí)間： 2014-01-24
上傳用戶：Thuan
for Masking Based SUBSPACE Speech Enhancement 一篇關(guān)于語(yǔ)音增強(qiáng)的 pdf ！值得共享
標(biāo)簽： Enhancement SUBSPACE Masking Speech
上傳時(shí)間： 2014-02-10
上傳用戶：李夢(mèng)晗
Optimal SUBSPACE-Based Signal Processing，美國(guó)The University of Rhode Island關(guān)于陣列信號(hào)處理中DOA的課件
標(biāo)簽： SUBSPACE-Based Processing Optimal Signal
上傳時(shí)間： 2016-10-25
上傳用戶：冇尾飛鉈
SUBSPACE Projection Based Blind Channel Order Estimation of MIMO Systems m file for a classical channel order estimation method
標(biāo)簽： Projection Estimation classical SUBSPACE
上傳時(shí)間： 2016-11-26
上傳用戶：weixiao99
A SOFT MODEL-ORDER SUBSPACE BASED SPEECH ENHANCEMENT ALGORITHM
標(biāo)簽： MODEL-ORDER ENHANCEMENT ALGORITHM SUBSPACE
上傳時(shí)間： 2016-12-07
上傳用戶：金宜
Description The MUSIC algorithm, proposed by Schmidt, first estimates a basis for the noise SUBSPACE and then determines the peaks the associated angles provide the DOA estimates. The MATLAB code for the MUSIC algorithm is sampled by creating an array of steering vectors corresponding to the angles in the vector angles.
標(biāo)簽： Description algorithm estimates proposed
上傳時(shí)間： 2013-12-08
上傳用戶：hgy9473
提出了一種改進(jìn)的LSM-ALSM子空間模式識(shí)別方法，將LSM的旋轉(zhuǎn)策略引入ALSM，使子空間之間互不關(guān)聯(lián)的情況得到改善，提高了ALSM對(duì)相似樣本的區(qū)分能力。討論中以性能函數(shù)代替經(jīng)驗(yàn)函數(shù)來(lái)確定拒識(shí)規(guī)則的參數(shù)，實(shí)現(xiàn)了識(shí)別率、誤識(shí)率與拒識(shí)率之間的最佳平衡；通過對(duì)有限字符集的實(shí)驗(yàn)結(jié)果表明，LSM-ALSM算法有效地改善了分類器的識(shí)別率和可靠性。關(guān) 鍵 詞 學(xué)習(xí)子空間; 性能函數(shù); 散布矩陣; 最小描述長(zhǎng)度在子空間模式識(shí)別方法中，一個(gè)線性子空間代表一個(gè)模式類別，該子空間由反映類別本質(zhì)的一組特征矢量張成，分類器根據(jù)輸入樣本在各子空間上的投影長(zhǎng)度將其歸為相應(yīng)的類別。典型的子空間算法有以下三種[1, 2]：CLAFIC(Class-feature Information Compression)算法以相關(guān)矩陣的部分特征向量來(lái)構(gòu)造子空間，實(shí)現(xiàn)了特征信息的壓縮，但對(duì)樣本的利用為一次性，不能根據(jù)分類結(jié)果進(jìn)行調(diào)整和學(xué)習(xí)，對(duì)樣本信息的利用不充分；學(xué)習(xí)子空間方法(Leaning SUBSPACE Method, LSM)通過旋轉(zhuǎn)子空間來(lái)拉大樣本所屬類別與最近鄰類別的距離，以此提高分類能力，但對(duì)樣本的訓(xùn)練順序敏感，同一樣本訓(xùn)練的順序不同對(duì)子空間構(gòu)造的影響就不同；平均學(xué)習(xí)子空間算法(Averaged Learning SUBSPACE Method, ALSM)是在迭代訓(xùn)練過程中，用錯(cuò)誤分類的樣本去調(diào)整散布矩陣，訓(xùn)練結(jié)果與樣本輸入順序無(wú)關(guān)，所有樣本平均參與訓(xùn)練，其不足之處是各模式的子空間之間相互獨(dú)立。針對(duì)以上問題，本文提出一種改進(jìn)的子空間模式識(shí)別方法。子空間模式識(shí)別的基本原理1.1 子空間的分類規(guī)則子空間模式識(shí)別方法的每一類別由一個(gè)子空間表示，子空間分類器的基本分類規(guī)則是按矢量在各子空間上的投影長(zhǎng)度大小，將樣本歸類到最大長(zhǎng)度所對(duì)應(yīng)的類別，在類x()iω的子空間上投影長(zhǎng)度的平方為()211,2,,()argmax()jMTkkjpg===Σx (1)式中 函數(shù)稱為分類函數(shù)；為子空間基矢量。兩類的分類情況如圖1所示。
上傳時(shí)間： 2013-12-25
上傳用戶：熊少鋒
Prony算法工具箱。Prony方法是用一組指數(shù)項(xiàng)的線性組合來(lái)擬和等間距采樣數(shù)據(jù)的方法,可以從中分析出信號(hào)的幅值、相位、阻尼因子、頻率等信息。considerations including data preprocessing, model order selection, model order selection criteria, signal SUBSPACE selection, signal and noise separation, root inspection and assessing residuals. The PTbox provides flexibility to compare and display analysis results simultaneously for several parameter variations.
標(biāo)簽： Prony considerations including data
上傳時(shí)間： 2015-09-11
上傳用戶：lizhizheng88
蟲蟲下載站版權(quán)所有 京ICP備2021023401號(hào)-1
