Gaussian Processes - Interactive Demo

This is an interactive implementation of a gaussian process written in javascript that runs in the browser. To use simply click on the main chart to add 'observations' and watch the model update its predictions! Three of the most common kernels are implemented: the radial basis function kernel, the linear kernel and the periodic kernel. It is possible to switch between them using the toggles above the chart. It is also possible to select multiple kernels - in this case the kernels are combined (using addition). Adjusting the kernel parameters will adjust both the kernel visualization and the main plot. I hope you enjoy seeing this vizualization!

RBF Kernel

$\sigma^2 \exp\left(-\frac{|{x-x^\prime}|^2}{2l^2}\right)$

Variance: $\sigma$=

Length: $l$= value:

Linear Kernel

$\sigma_b^2 + \sigma^2 (x - c)(x^\prime - c)$

Variance: $\sigma$ =

Variance: $\sigma_b$ =

Offset: $c$ = value:

Periodic Kernel

$\sigma^2 \exp \left(-\frac{2}{\ell^2}\sin^2 \left( \pi \frac{\lvert x_a - x_b \rvert}{p}\right) \right)$

Variance: $\sigma$ =

Length: $l$ =

Periodicity: $p$ = value: