Stanford machine learning course summary

Preface

First, thank you. Wu enda establish Coursera Such an excellent online learning platform , And what he posted on this platform machine learning Course .

This course will the basic knowledge of the whole field of machine learning , In an easy way , In simple terms . So that a student with high school mathematics knowledge can understand .

If you want to get involved in machine learning 、 The field of artificial intelligence , Or have a strong interest in this field and want to know more about , Then you will find many materials recommended by the introductory course of machine learning , There are a series of courses by teacher Wu Enda . Even in most materials , Put this course in the preferred position .

therefore , I organized teacher Wu Enda's courses into MarkDown The format of , It's easy to check and learn . The following is the table of contents of specific chapters , Each article has a corresponding video connection address ：

Catalog

• The first week
  • Welcome to machine learning
  • Supervised learning
  • Unsupervised learning
  • Univariate linear regression
  • Parameter learning - Gradient descent algorithm
  • Review of linear algebra
• In the second week of
  • Programming environment settings -Octave:MATLAB
  • Multiple linear regression analysis
  • Calculation and analysis of parameters
  • Octave/Matlab Instructions
• The third week
  • Classification and expression
    • classification
    • Suppose the function expression
    • Decision boundaries
  • Logistic The regression model
    • The cost function of logical regression
    • Simplify the cost function and gradient descent
    • Advanced optimization
  • Multi category classification problem ： One to many
  • Regularization ： Solve the over fitting problem
    • Solve the over fitting problem
    • Cost function
    • Regularize linear regression
• The fourth week
  • Neural network is introduced
    • Nonlinear hypothesis
    • Neural networks and the brain
  • neural network
  • Examples of neural network applications
• Week 5
  • Training neural network
    • Cost function
    • Back propagation (B-P)
    • Intuitive introduction of back propagation algorithm
  • BP Algorithm
  • Neural network realizes automatic driving
• Week six
  • Evaluate a learning algorithm
    • How to avoid detours ？
    • Evaluation hypothesis function
    • Selection of polynomial model and training set / Verification set / Partition of test set
  • deviation VS variance
    • deviation VS variance
    • Regularization and deviation / variance
    • The learning curve (Learning Curves)
    • Review and decide what to do next
  • Machine learning system design
    • Build a spam classifier
    • error analysis
  • Operation skew data
    • Error metrics for offset classes
    • Precision and recall exercises
  • Use big data sets
• Week 7
  • Large spacing classification SVM
    • Optimization objectives
    • Big gap intuition
    • The mathematics behind the large spacing classifier ( Selected studies ))
  • Kernel function
  • Use SVM
• Week eight
  • clustering
    • Introduction to unsupervised learning
    • K-Means Algorithm
    • Optimization objectives
    • Random initialization
    • Select the number of clusters
  • PCA Dimension reduction
• Week 9
  • Density estimation & Anomaly detection
  • Build an anomaly detection system
  • Multivariate Gaussian distribution （ Selected studies ）
  • Predict movie ratings
    • Predict movie ratings
    • Collaborative filtering
    • Low rank matrix decomposition
• Week 10
  • Gradient descent of large data sets
    • Learning algorithm for processing big data
    • Stochastic gradient descent
    • Small batch gradient descent
    • Convergence of random gradient descent
  • Advanced themes
    • Online learning
    • Map Reduce Parallel with data
• Week 11
  • Photo OCR
    • Problem description and pipeline （Pipeline）
    • Slide form
    • Get a lot of data and human data
    • Limit analysis ： What's the next step on the assembly line