liveBook · Manning

Kernel function

A kernel function is a mathematical function that measures similarity between inputs, often by representing an inner product in an implicit feature space.

How 6 Manning books explain it

Grokking Machine Learning

A mathematical function that represents an inner product in a transformed feature space, allowing a model to learn nonlinear decision boundaries.

How it is used
Uses kernels with SVMs to classify datasets that lack a linear decision boundary.
How deep it goes
Explains similarity, feature mapping, and RBF and polynomial examples, while deferring the full theory of kernels and feature maps.
Examples
Uses XOR, circular and one-dimensional datasets to illustrate polynomial and RBF kernel classification.
What you can do after
Apply via SVC by selecting a polynomial or RBF kernel and setting its degree or gamma.

Read how Grokking Machine Learning explains kernel function →

Timeless Algorithms, The Foundational Ideas

A function that computes an inner product in a feature space, allowing some learning methods to operate implicitly in that space.

How it is used
Extends support vector machines to nonlinear classification through implicit feature-space comparisons.
How deep it goes
Derives the kernel substitution in the support-vector decision function and compares kernel families, parameters, and model-selection tradeoffs.
Examples
Calculates a transformed temperature dot product and compares linear, polynomial, and RBF boundaries on synthetic two-class data.
What you can do after
Evaluate kernel choices and parameter settings using their boundary behavior and cross-validation rather than training accuracy alone.

Read how Timeless Algorithms, The Foundational Ideas explains kernel function →

Causal Inference for Data Science

A kernel is a weighting function that assigns observation weights according to distance from a target point.

How it is used
Weights nearby observations more heavily when estimating local effects in regression discontinuity designs.
How deep it goes
Defines two kernels, shows bandwidth transformation, and defers elaborate bandwidth selection to package documentation.
Examples
Uses uniform and triangular functions, kernel figures, and local estimation around a time cutoff.
What you can do after
Evaluate how kernel choice and bandwidth control which observations influence a local regression estimate.

Read how Causal Inference for Data Science explains kernel function →

Experimentation for Engineers

A kernel function assigns a similarity or weight to an observation based on its distance from another input.

How it is used
Weights nearby measurements when Gaussian process regression estimates CPU time in Bayesian optimization.
How deep it goes
Defines distance-based weighting, implements squared exponential kernels and normalization, but defers leave-one-out cross-validation details.
Examples
Uses CPU-time measurements at parameters 0.5 and 0.0, evaluates queries 0.25 and 0.4, and plots surrogate curves.
What you can do after
Apply it via NumPy to normalize distance-based weights and estimate query values from measured parameters.

Read how Experimentation for Engineers explains kernel function →

Build AI Drug Discovery Pipelines

A kernel function is a smooth, symmetric, non-negative weighting function that determines how an observation contributes to an estimated density at a location.

How it is used
Uses weighted contributions to estimate continuous distributions for pharmacophore pair distances.
How deep it goes
Defines kernel shape and influence, then applies Gaussian KDE with cross-validated bandwidth selection.
Examples
Uses Gaussian and cosine kernels, bimodal data, and fragment-ensemble pharmacophore pair distances.
What you can do after
Apply it via scikit-learn to fit KDE models and estimate pharmacophore distance distributions.

Read how Build AI Drug Discovery Pipelines explains kernel function →

Machine Learning Algorithms in Depth

A positive-definite function that measures similarity between pairs of inputs and defines an associated feature-space inner product.

How it is used
Supplies pairwise similarities in an SVM dual and covariance between function values in Gaussian process regression.
How deep it goes
Connects kernels to SVM dual optimization and GP posterior conditioning, then implements RBF evaluations without explicitly constructing feature vectors.
Examples
Uses RBF kernel code for an SVM and GP regression, alongside graph, string, and document kernels as other similarity measures.
What you can do after
Build it from scratch in an SVM optimized with CVXOPT or in Gaussian process regression using NumPy and SciPy.

Read how Machine Learning Algorithms in Depth explains kernel function →

Also covered in

What to know first

Related concepts

Questions about kernel function

What is kernel function?

A kernel function is a mathematical function that measures similarity between inputs, often by representing an inner product in an implicit feature space.

Which Manning books explain kernel function?

Grokking Machine Learning, Timeless Algorithms, The Foundational Ideas, Causal Inference for Data Science, Experimentation for Engineers, Build AI Drug Discovery Pipelines and Machine Learning Algorithms in Depth.

What should I know before learning kernel function?

Manning books that teach kernel function usually introduce covariance, inner product, linear regression and machine learning first.

What kinds of kernel function are there?

Kinds of kernel function include radial basis function kernel and polynomial kernel.

What is kernel function a kind of?

Kernel function is a kind of similarity metric. A similarity metric is a function that quantifies how alike two objects or representations are according to specified properties.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.