Gu's Blog

In me the tiger sniffs the rose

Training a Wasserstein GAN on the free google colab TPU

使用谷歌免费TPU训练Wasserstein GAN 模型

The jupyter notebook is available on my github repo. Click HERE to play with google colab on live! Why using TPU? TPU is much faster than GPU. A single TPU contains 8 cores, each with ...

EM Algorithm and Gaussian Mixture Model for Clustering

EM算法与高斯混合模型

In the last post on EM algorithm, we introduced the deduction of the EM algorithm and use it to solve the MLE of the heads probability of two coins. In this post, we will apply EM algorithm to more...

An Introduction to Expectation-Maximization (EM) Algorithm

EM算法详解

Expectation Maximization (EM) algorithm is a special case of MLE where the observations (data samples ) are inherently related with some hidden variables (). First of all, we need to review the bas...

An Introduction to Support Vector Machines (SVM): A Python Implementation

支持向量机(SVM)概述:Python 代码实现

The Jupyter notebook is available at my github: SupportVectorMachine/SVM-Tutorial First of all, we need to implement the SVM solver based on the SMO algorithm. 1 2 3 4 import numpy as np import m...

An Introduction to Support Vector Machines (SVM): Sequential Minimal Optimization (SMO)

支持向量机(SVM)概述:SMO算法求解对偶问题

Recall the Kernel SVM dual problem: Dual Problem $$ \max_{\lambda, \mu} L(\lambda)= \sum_{i=1}^{n}\lambda_i - \frac{1}{2}\sum_{i,j}\lambda_i \lambda_j y_i y_j K_{i,j} \\ \begin{align} s.t.\ ...

An Introduction to Support Vector Machines (SVM): kernel functions

支持向量机(SVM)概述:核函数

Recall of the Slack SVM dual problem: Dual Problem $$ \max_{\lambda, \mu} \sum_{i=1}^{n}\lambda_i - \frac{1}{2}\sum_{i,j}\lambda_i \lambda_j y_i y_j \mathbf{x}_i^T\mathbf{x}_j\\ \begin{align} ...

An Introduction to Support Vector Machines (SVM): SVM with slack variables

支持向量机(SVM)概述:具有松弛变量的支持向量机

Just to clarify, these contents are mainly summarized from the course I took: “Fundamental of Big Data Analytics”, taught by Prof. Mathar Rudolf. For for information please visit: https://www.ti...

An Introduction to Support Vector Machines (SVM): Dual problem solution using GD

支持向量机(SVM)概述:使用梯度下降求解对偶问题

Just to clarify, these contents are mainly summarized from the course I took: “Fundamental of Big Data Analytics”, taught by Prof. Mathar Rudolf. For for information please visit: https://www.ti...

An Introduction to Support Vector Machines (SVM): Convex Optimization and Lagrangian Duality Principle

支持向量机(SVM)概述:凸优化与拉格朗日对偶问题

Just to clarify, these contents are mainly summarized from the course I took: “Fundamental of Big Data Analytics”, taught by Prof. Mathar Rudolf. For for information please visit: https://www.ti...

An Introduction to Support Vector Machines (SVM): Gradient Descent Solution

支持向量机(SVM)概述:梯度下降法

Just to clarify, these contents are mainly summarized from the course I took: “Fundamental of Big Data Analytics”, taught by Prof. Mathar Rudolf. For for information please visit: https://www.t...