Machine Learning

We also have models that walk around with the dataset in their carry-on. These are models such as:

1. Decision Trees
2. SVM
3. Nonparametric Regressions: K-nearest neighbours, Locally Weighted
4. Random Forests

This accounts for about 60% of the Machine Learning Methods we have.

By definition a parametric model is one that has fixed parameters to learn, i.e. weights in Linear Regression: $w_0, w_1, ..., w_n$. Conversely, a non-parametric model does not have a fixed number of parameters to learn: K-means clustering for example just clusters the data as best as it can.

We can list some more models:

1. Linear Regression
2. Ridge Regression
3. Lasso Regression
4. Logistic Regression
5. Neural Networks
6. Perceptron
7. Naive Bayes