mfccccc

粒子滤波器程序，应用了SIS方法和蒙特卡罗方法(particle filter codes，using SIS method and Monte Carlo method)

利用jacobi法求解病态线性方程，其中矩阵为hilbort矩阵(Jacobi method used to solve ill-linear equations, including matrix hilbort matrix)

利用MATLAB编写GMM图像处理模块。具有较好的效果。。。(GMM prepared using MATLAB image processing module. With good results. . .)

说明：  TIMIT语音数据库,希望给语音研究者提供帮助。(TIMIT speech database, I want to give help to the researchers of voice .)

PCA+SVM用于人脸识别，，很好用，，有详细注释(PCA+SVM for face recognition, and useful, detailed notes)

matlab相關教學書籍matlab相關教學書籍matlab相關教學書籍(proficient MATLAB)

数学交互界面，可以实现求解线性方程组，三维绘图，二重积分等(Math interactive interface for solving linear equations, three-dimensional drawing, the double integral)

I have uploaded complete GUI code for Aircraft A320 control

深度神经网络在函数近似中表现优越，然而需要从头开始训练。另一方面，贝叶斯方法，像高斯过程（GPs），可以利用利用先验知识在测试阶段进行快速推理。然而，高斯过程的计算量很大，也很难设计出合适的先验。本篇论文中我们提出了一种神经模型，条件神经过程（CNPs），可以结合这两者的优点。CNPs受灵活的随机过程的启发，比如GPs，但是结构是神经网络，并且通过梯度下降训练。CNPs通过很少的数据训练后就可以进行准确的预测，然后扩展到复杂函数和大数据集。我们证明了这个方法在一些典型的机器学习任务上面的的表现和功能，比如回归，分类和图像补全(Deep neural networks perform well in function approximation, but they need to be trained from scratch. On the other hand, Bayesian methods, such as Gauss Process (GPs), can make use of prior knowledge to conduct rapid reasoning in the testing stage. However, the calculation of Gauss process is very heavy, and it is difficult to design a suitable priori. In this paper, we propose a neural model, conditional neural processes (CNPs), which can combine the advantages of both. CNPs are inspired by flexible stochastic processes, such as GPs, but are structured as neural networks and trained by gradient descent. CNPs can predict accurately with very little data training, and then extend to complex functions and large data sets. We demonstrate the performance and functions of this method on some typical machine learning tasks, such as regression, classification and image completion.)

In MATLAB, usually we use function bvp4c() to solve boundary value problem (BVP) of ODE. However, we also can use function ode45() to solve BVP of ODE too. However, ode45() needs all initial values at one point. Therefore we have to construct an auxiliary function, whose input argument is the missing initial value and the return output is the given boundary value, Then we can call function fsolve() to get the missing initial value. This method is particularly efficient it the ODE has one or more parameters. Also this method is essentilal for GNU Octave, because whose function bvp4c() is not programmed for parameter ODE yet.