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.
The variable that you try to predict is continuous.
- predicting house price
The variable that you try to predict is discreet rather than continuous.
- predicting whether or not a certain tumor is malignant
How and why learning algorithms work?
- 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
- Understand gene data
- Computer vision / image processing
- Organize computing clusters
- Social network analysis
- Market segmentation
- 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?
Make a sequence of good decisions over time.
- Auto driving
- Web crawling