Unit: Machine Learning
Lesson: Linear Regression
Understanding Linear Regression
How Does Linear Regression Work?
The Least Squares Method
Gradient Descent for Linear Regression (Optional)
Numerical Example
Extension to Multiple Features (Optional)
Linear Regression Notebook (Optional)
Linear Regression Quiz
Carl Friedrich Gauss in 1795, to study planetary and stellar movements
Isaac Newton, to study gravity
Albert Einstein, to develop relativity theory
Thomas Bayes, to develop probability theory
y must be either -1 or +1
y must be a whole number
y must be positive
y can be any real number
The slope of the line
The predicted score for 0 hours of study
The number of study hours
The average test score
To give less weight to smaller errors
To give more weight to larger errors
To make all errors positive, regardless of whether we predicted too high or too low
To match the scale of the original data