Back to Math Foundry

Guided curriculum

Your learning journey

Follow all seven stages in order, or use your latest knowledge check to start at a focused recommendation. Nothing is locked.

Current recommendation

Linear equations stay balanced

Mathematical language · 10 min

Continue
0 of 48 complete0%
1

Mathematical language

0/9

Build the algebra, vectors, matrices, and calculus used throughout data science.

  1. Linear equations stay balanced 10 min · introNext
  2. Functions are machines 11 min · intro
  3. Slope measures change 12 min · intro
  4. Exponents, logs & growth 13 min · intro
  5. Dot product, norms, and similarity 10 min · core
  6. Matrices transform space 8 min · intro
  7. Derivatives: change right now 14 min · intro
  8. Integrals accumulate 14 min · intro
  9. Series, approximation & Big-O 15 min · intro
2

Linear algebra decompositions

0/3

Understand the structure behind dimensionality reduction and representation learning.

  1. Eigenvalues and eigenvectors 11 min · core
  2. Singular value decomposition 12 min · core
  3. Principal component analysis 12 min · core
3

Probability and statistical inference

0/10

Learn how uncertainty, estimation, testing, and experiments fit together.

  1. Common distributions 12 min · core
  2. Bayes' theorem 10 min · core
  3. Maximum likelihood estimation 11 min · core
  4. CLT and sampling distributions 10 min · core
  5. Confidence intervals 9 min · core
  6. Hypothesis testing and p-values 12 min · core
  7. Bootstrap and resampling 10 min · core
  8. Correlation vs causation 11 min · core
  9. Experiment design and power 12 min · core
  10. A/B testing pitfalls 12 min · advanced
4

Machine-learning foundations

0/9

Connect preprocessing, linear models, generalization, and evaluation.

  1. Feature scaling and preprocessing 8 min · intro
  2. Linear regression 12 min · core
  3. Logistic regression 11 min · core
  4. Classification metrics 11 min · core
  5. Bias-variance tradeoff 12 min · core
  6. Overfitting and generalization 9 min · core
  7. L1 and L2 regularization 11 min · core
  8. Cross-validation 9 min · core
  9. Class imbalance 10 min · core
5

Model families

0/5

Compare the major supervised and unsupervised approaches and when to use them.

  1. KNN & Naive Bayes 12 min · core
  2. SVMs & the Kernel Trick 12 min · core
  3. Decision trees and random forests 12 min · core
  4. Gradient boosting 12 min · advanced
  5. Clustering 12 min · core
6

Deep-learning foundations

0/6

Follow information and gradients through a neural network before adding architectures.

  1. Forward pass 8 min · intro
  2. Activation functions 9 min · core
  3. Backpropagation 12 min · core
  4. Optimizers: SGD, momentum, Adam 11 min · core
  5. Batch and layer normalization 12 min · core
  6. Dropout and regularization 10 min · core
7

Modern deep learning

0/6

Progress from learned representations to attention, transformers, and transfer learning.

  1. Embeddings 9 min · core
  2. Convolutional neural networks 11 min · core
  3. RNNs & LSTMs 12 min · core
  4. Attention mechanism 12 min · core
  5. Transformer architecture 12 min · advanced
  6. Transfer learning and fine-tuning 10 min · core