All Modules

Machine Learning

Every machine learning idea is a mathematical object. Through 29 interactive demonstrations across loss landscapes, eigendecompositions, kernels, automatic differentiation, Bayesian inference, information theory, equivariance, and manifolds, see the math that quietly runs every model.

See also: Linear Algebra for eigenvectors and SVD, Optimization for the foundations of gradient methods, and Probability for the priors and likelihoods behind Bayesian methods.