ML Basic

Applications of ML

Text Recognition;
Spam Filtering;
Recommendation System;
Medical Diagnosis;
Defeat Human;
Self-driving Car.

Types

Supervised learning: learning by examples
Labeled data;
Direct feedback;
Predict outcome/future’

Unsupervised learning: discovering patterns itself
No labels;
No feedback;
Find hidden structure.

Reinforcement learning: feedback right/wrong
Decision process;
Reward system;
Learn series of actions.

Types of tasks of ML

Classification:
Map an element to a category
Regression:
Predict numerical values
Similarity:
Find more like this
Ranking:
Return an ordered list(based on relevance)
Sequence Prediction:
Predict the next one in the series

Regression vs. Classification

1 A regression model predicts continuous values.
e.g.: What is the value of a house in California?
What is the probability that a user will click on this ads?

2 A classification model predicts discrete values.
e.g.: Is a given email message spam or not spam?
Is this an image of a dog, a cat, or a hamster?