Nonlinear classifier

As explained in Grokking Machine Learning

A classifier whose decision boundary is not representable as a single affine linear function of the input features.

A nonlinear classifier separates classes with a boundary that cannot be represented by one linear boundary. In low dimensions, linear boundaries look like lines or planes; nonlinear ones can form complicated curves or surfaces. section 10.0 section 11.0

Neural networks create nonlinear classifiers by composing linear classifiers. For datasets that need highly nonlinear boundaries, the kernel method can be used with a support vector machine (SVM). section 10.0 section 11.0

Figure 10.1 A neural network. It may look complicated, but in the next few pages, we will demystify this image.

A neural network diagram offers a visual starting point for understanding how combined building blocks can produce a nonlinear classifier. section 10.0

How Grokking Machine Learning covers it

How deep it goes

Defines nonlinear boundaries geometrically and links them to compositions of linear classifiers.

Examples

Contrasts lines and planes with complicated curves and surfaces as possible decision boundaries.

What it assumes

Builds on perceptrons and linear classifiers, while recalling discrete and continuous perceptrons.

What you can do after

Apply it via the kernel method with an SVM to classify datasets using highly nonlinear boundaries.

How it connects

Connects nonlinear classifiers with neural networks, perceptrons, linear classifiers, support vector machines, and the kernel method.

liveBookmanning.com

Read Grokking Machine Learning on liveBook Get the book at manning.com

Related concepts

• language modelproduced by · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
• kernel methodproduced by · A kernel method is a machine-learning technique that uses a kernel function to model relationships in a feature space without explicitly computing the feature representations.

This explainer was generated by Manning's knowledge graph from the text of Grokking Machine Learning. The section links point to the parts of the book it draws on.