Class Notes – Machine Learning, Lecture 1

Subject : Machine Learning (CS 229)
Instructor : Prof. Andrew Ng

machine learning

• come as part of Artificial Intelligence research
• Used to solve problem which difficult to be manually programmed
• Used for data mining

Indirect use of Machine Learning on our daily life :

• US Mail – to read zip code on envelope
• Check – to read money number
• Amazon, ebay, netflix – Suggest product
• Understanding Human Genome

prerequisite, know about:

• Big O
• Basic Data structure (Queue, Stack, Binary search)
• programming
• Basic probability and statistics
• Basic Linear Algebra especially using matrix

Machine Learning Definition :

• Field of Study that gives computer the ability to learn without being explicitly programmed (Arthur Samuel, 1959)
• Well Posed learning problem. A Computer program is said to learn from experience E with some respect to some task T and some performance measure P if its performance on T, as measured by P improves with experience E

Machine Learning Classification :

• Supervised Learning : supervise the algorithm by providing data to it.
• Learning Theory
• Unsupervised Learning :  give data to algorithm and let them find interesting structure from data
• Reinforced Learning : Give algoritm reward when it right and punishment when it wrong. Application Sample : autopilot helicopter

Application of Unsupervised Learning

• Organize Computer Cluster
• Social Network Analysis
• Market Segmentation
• Astronomical Data analysis
• Computer Vision / Computer image processing (build 3D view from single still image)

Cocktail party problem : lots of people talking to each other how to just hear from people that you want to hear or in the other way filtering noises.