# CS229 Note Lecture 01

## Machine Learning Definition

Arthur Samuel (1959). Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed.

Tom Mitchell (1998) Well-posed Learning Problem: A computer program is said to *learn* from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.

## Supervised Learning

### Regression

The variable that you try to predict is continuous.

**EXP**

- predicting house price

### Classification

The variable that you try to predict is discreet rather than continuous.

**EXP**

- predicting whether or not a certain tumor is malignant

## Learning Theory

How and why learning algorithms work?

**EXP**

- Proof theorems on when you can guarantee that a learning algorithm will work
- Find what algorithms can approximate different functions well
- How much train data do we need

## Unsupervised Learning

### Clustering

**EXP**

- Understand gene data
- Computer vision / image processing
- Organize computing clusters
- Social network analysis
- Market segmentation
- Astronomy
- Cocktail party problem (independent component analysis)

Is that all of these people talking, can you separate out the voice of just the person you are interested in talking to with all this loud background noise?

## Reinforcement Learning

Make a sequence of good decisions over time.

**EXP**

- Auto driving
- Robotics
- Web crawling