Posted on Categories:CS代写, Machine Learning, 机器学习, 计算机代写

# 机器学习代考_Machine Learning代考_COMP5318 Density Estimation

avatest™

## avatest™帮您通过考试

avatest™的各个学科专家已帮了学生顺利通过达上千场考试。我们保证您快速准时完成各时长和类型的考试，包括in class、take home、online、proctor。写手整理各样的资源来或按照您学校的资料教您，创造模拟试题，提供所有的问题例子，以保证您在真实考试中取得的通过率是85%以上。如果您有即将到来的每周、季考、期中或期末考试，我们都能帮助您！

•最快12小时交付

•200+ 英语母语导师

•70分以下全额退款

## 机器学习代考_Machine Learning代考_Density Estimation

As we have seen in the discussion of the statistical data-modeling procedure, before we can apply the plug-in MAP decision rule, the fundamental problem is how to estimate the unknown data distribution based on a finite set of training samples that are presumably drawn from this distribution. This corresponds to a standard problem in statistics, namely, density estimation. As we have seen, we normally take the so-called parametric approach to this problem. In other words, we first choose some parametric probabilistic models, and then the associated parameters are estimated from the finite set of training samples. The advantage of this approach is that we can convert an extremely challenging problem of density estimation into a relatively simple parameter-estimation problem. By estimating the parameters, we find the best fit to the unknown data distribution in the family of some prespecified generative models. Similar to discriminative models, parameter estimation for generative models can also be formulated as a standard optimization problem. The major difference here is that we need to rely on different criteria to construct the objective function for generative models. In the following, we will explore the most popular method for parametric density estimation, namely, maximum-likelihood estimation (MLE).

## 机器学习代考_Machine Learning代考_Maximum-Likelihood Estimation

Assume that we are interested in estimating an unknown data distribution $p(\mathbf{x})$ based on some samples randomly drawn out of this distribution; that is, $\mathscr{D}N=\left{\mathbf{x}_1, \mathbf{x}_2 \cdots, \mathbf{x}_N\right}$, where each sample $\mathbf{x}_i \sim p(\mathbf{x})(\forall i=1,2 \cdots, N)$. An important assumption in density estimation is that we assume these samples are independent and identically distributed (i.i.d.), which means that all these samples are drawn from the same probability distribution, and all of them are mutually independent. As we will see later, the i.i.d. assumption will significantly simplify the parameter-estimation problem in density estimation. In a parametric density-estimation method, we first choose a probabilistic model, $\hat{p}{\boldsymbol{\theta}}(\mathbf{x})$, to approximate this unknown distribution $p(\mathbf{x})$, where $\boldsymbol{\theta}$ denotes the parameters of the chosen model. The unknown model parameters $\theta$ are then estimated from the collected training samples $\mathscr{D}_N$. The most popular method for this parameter estimation problem is the so-called MLE. The basic idea of MLE is to estimate the unknown parameters $\theta$ by maximizing the joint probability of observing all training samples in $D_N$ based on the presumed probabilistic model. That is,

\begin{aligned} \boldsymbol{\theta}{\mathrm{MLE}} &=\arg \max {\boldsymbol{\theta}} \hat{p}{\boldsymbol{\theta}}\left(\mathscr{D}_N\right) \ &=\arg \max {\boldsymbol{\theta}} \hat{p}{\boldsymbol{\theta}}\left(\mathbf{x}_1, \mathbf{x}_2, \cdots, \mathbf{x}_N\right) \ &=\arg \max {\boldsymbol{\theta}} \prod_{i=1}^N \hat{p}_{\boldsymbol{\theta}}\left(\mathbf{x}_i\right) . \end{aligned}

## MATLAB代写

MATLAB 是一种用于技术计算的高性能语言。它将计算、可视化和编程集成在一个易于使用的环境中，其中问题和解决方案以熟悉的数学符号表示。典型用途包括：数学和计算算法开发建模、仿真和原型制作数据分析、探索和可视化科学和工程图形应用程序开发，包括图形用户界面构建MATLAB 是一个交互式系统，其基本数据元素是一个不需要维度的数组。这使您可以解决许多技术计算问题，尤其是那些具有矩阵和向量公式的问题，而只需用 C 或 Fortran 等标量非交互式语言编写程序所需的时间的一小部分。MATLAB 名称代表矩阵实验室。MATLAB 最初的编写目的是提供对由 LINPACK 和 EISPACK 项目开发的矩阵软件的轻松访问，这两个项目共同代表了矩阵计算软件的最新技术。MATLAB 经过多年的发展，得到了许多用户的投入。在大学环境中，它是数学、工程和科学入门和高级课程的标准教学工具。在工业领域，MATLAB 是高效研究、开发和分析的首选工具。MATLAB 具有一系列称为工具箱的特定于应用程序的解决方案。对于大多数 MATLAB 用户来说非常重要，工具箱允许您学习应用专业技术。工具箱是 MATLAB 函数（M 文件）的综合集合，可扩展 MATLAB 环境以解决特定类别的问题。可用工具箱的领域包括信号处理、控制系统、神经网络、模糊逻辑、小波、仿真等。