Parametric model vs Nonparametric model

・In Japanese

■What is a parametric model?

A parametric model is a model in which parameters are set after determining the shape of the function. If there is a lot of data, fitting can be performed with good accuracy. But if there is little data, it is not possible to know what kind of model to use, so the accuracy is not so good.



Example of a parametric model is logistic regression.

■What is a nonparametric model?

A nonparametric model is a model that uses a method of fitting data without determining the type of function in advance. The model type allows you to select the best model for each data, so you can get a good fitting.

Examples of nonparametric models are support vector machines and Gaussian process regression.

List of related articles TOP MENU Machine learning ・ε-greedy method ・Reinforcement learning ・Temporal Difference learning ・Regression vs Classification vs Clustering ・Parametric vs Nonparametric model ・LSTM ・Error function ・Experience Replay