Regression
Regression is a way for a computer to make an educated guess about a specific number, like predicting the future price of a house based on its size and location.
A supervised machine learning task where a model is trained to map input features to a continuous numerical output, typically evaluated using metrics like Mean Squared Error.
A class of supervised learning algorithms designed to model the functional relationship between input variables and a continuous target variable, effectively approximating a mapping function f(x) = y where y belongs to the set of real numbers.
evolution
- 1805 · historyMethod of Least Squares
Adrien-Marie Legendre published the first formal method for minimizing the sum of squared residuals to fit a line to data.
- 1877 · historyRegression toward the Mean
Francis Galton coined the term 'regression' while observing that extreme biological traits tend to return to the average in subsequent generations.
- 1950 · historyStatistical Learning Theory
The formalization of regression as a predictive modeling task became a cornerstone of early machine learning and statistical inference.
- 1990 · historyGeneralized Additive Models
The integration of flexible, non-linear regression techniques allowed models to capture complex patterns in high-dimensional data.
- 2012 · historyDeep Learning Regression
The success of AlexNet demonstrated that deep neural networks could be used for high-precision regression tasks in computer vision and beyond.