Analysis

Learn Mathematics of Machine Learning

The math behind machine learning: the vector/matrix data model, matrix calculus and Jacobians, backpropagation as the chain rule, gradient descent and its convergence, stochastic gradient descent, loss functions (MSE, cross-entropy, softmax), regularization, PCA via SVD, the kernel trick, the bias–variance tradeoff, and the probabilistic (maximum-likelihood) view of regression.

Free to start · adaptive placement finds your level · reviews timed so it stays learned.

What you'll learn

14 lessons in Mathematics of Machine Learning

Vectors, matrices & the ML data modelMatrix calculus I — gradients & JacobiansMatrix calculus II — the chain rule for vector functionsBackpropagation as the chain ruleGradient descent for learningConvexity & the global-minimum guaranteeConvergence of gradient descentStochastic gradient descent & mini-batchesLoss functions — MSE, cross-entropy & softmaxRegularization — L2, L1 & the bias they addPCA via SVD — dimensionality reductionThe kernel trick & feature mapsBias–variance tradeoff & generalizationProbabilistic view — MLE for regression
How Erudia teaches

Built to be understood — and remembered.

Every idea is taught with motivation and a worked example before the drills, and an FSRS spaced-repetition engine schedules each review for the moment just before you'd forget it. A short placement check finds what you already know, so you start Mathematics of Machine Learning exactly where it's useful.

Related Analysis subjects