Machine Learning

Machine learning is the science of getting computers to work without being explicitly programmed. The Machine learning course offered by Stanford University provides an introduction to this. It flows very smoothly and in the way that the hardest and toughest topics get absorbed with ease. Indeed, the instructor has delivered and designed an incredible course. For starting ML, no other resource could be better,

I.  Introduction

    - Introduction

    - Model and Cost function

    - Parameter learning

    - Linear algebra review

II. Linear regression with multiple variables 

    - Multiple linear regression

    - Computing parameters analytically

III. Logistic regression

    - Classification and representation

    - Logistic regression model

    - Multiclass classification

    - Solving the problem of overfitting

IV. Neural networks

    - Neural networks

    - Applications

V. Neural networks: Learning

    - Cost function and backpropagation

    - Backpropagation in practice

    - Application of neural networks

VI. Applying advice for ML

    - Evaluation of a learning algorithm

    - Bias vs variance

    - Building spam classifier

    - Handling skewed data

    - Using large data-sets

VII. Support Vector Machines

    - Large margin classification

    - Kernels

    - SVM in practice

VIII. Unsupervised learning

    - Clustering

    - Principal component analysis

    - Applying PCA

IX. Anomaly Detection & Recommender Systems

    - Anomaly detection

    - Density estimation

    - Building anomaly detection system

    - Recommender systems

    - Predicting movie ratings

    - Collaborative filtering

    - Low rank matrix factorization

X. Large Scale ML

    - Gradient descent with large datasets

    -  Advanced topics

XI. Application: photo OCR

    - Photo OCR

Click here to read notes

Click here to read answers to quizzes and assignments