ML- Unsupervised Learning, K-means, Dimentionality Reduction
Posted mashuai-191
tags:
篇首语：本文由小常识网(cha138.com)小编为大家整理，主要介绍了ML- Unsupervised Learning, K-means, Dimentionality Reduction相关的知识，希望对你有一定的参考价值。
Clustering
K-means:
基本思想是先随机选择要分类数目的点，然后找出距离这些点最近的training data 着色，距离哪个点近就算哪种类型，再对每种分类算出平均值，把中心点移动到平均值处，重复着色算平均值，直到分类成功.
One way to choose K is elbow method
Dimentionality Reduction: to save space of memory and speed up compute. 还有一个作用是可以用降维来visualize data.
降维最常用的算法PCA (Principal Component Analysis)
the 1st step of PCA algo is data preprocessing
PCA algo in matlab:
How to de-compress back from 100-dimentional to 1000-dimentional
How to choose the parameter K
Advice for using PCA. PCA is often used for data compresion and visualization. it is bad to use it to prevent overfitting.
以上是关于ML- Unsupervised Learning, K-means, Dimentionality Reduction的主要内容，如果未能解决你的问题，请参考以下文章
Unsupervised Deep Learning – ICLR 2017 Discoveries
CVICML2015_Unsupervised Learning of Video Representations using LSTMs
CVICCV2015_Unsupervised Learning of Visual Representations using Videos
Machine Learning——Unsupervised Learning（机器学习之非监督学习）
如何区分监督学习(supervised learning)和非监督学习(unsupervised learning)
paper 124：转载无监督特征学习——Unsupervised feature learning and deep learning