Learn AI Create the Future — with Pranay
Main site

All sections

The whole curriculum, grouped into 14 areas. Start at the top if you are new. If you came here for one thing, filter to its area or search — everything is on this page or one click from it.

Every lesson here is written by .

14 areas 117 sections 1,162 of 1,167 lessons published 10 new lessons every day

Showing Statistics and Experiments — 4 sections, 42 lessons. Show all 14 areas

Statistics and Experiments

Telling a real result from a lucky one: significance, A/B tests, and working out what actually caused what.

4 sections 42 of 42 lessons published

Hypothesis Testing and Inference

The statistics that decide whether a difference you can see is a difference that is really there.

11 published

  1. Hypothesis testing
  2. T-tests
  3. The chi-squared test
  4. ANOVA
  5. Non-parametric tests
  6. Pearson, Spearman and Kendall
  7. Confidence intervals
  8. The central limit theorem
  9. Statistical power and sample size
  10. Multiple testing correction
  11. Q-Q plots and normality checks

Resampling, Likelihood and Bayes

Answering statistical questions by simulation and by belief updating, when the formula does not exist or you do not trust it.

8 published

  1. The bootstrap
  2. Permutation tests
  3. Monte Carlo simulation
  4. Maximum likelihood estimation
  5. Bayesian vs frequentist thinking
  6. Priors, posteriors and conjugate updating
  7. MCMC from scratch
  8. Kernel density estimation

Experiment Design and A/B Testing

Running an online experiment that gives you an answer you can defend, and spotting the ways it goes wrong.

10 published

  1. Randomisation units and assignment
  2. A/A tests
  3. Peeking and sequential testing
  4. Sample ratio mismatch
  5. CUPED and variance reduction
  6. Novelty and primacy effects
  7. Interference and network effects
  8. Switchback experiments
  9. Guardrail metrics
  10. Bayesian A/B testing

Causal Inference Basics

Working out what actually causes what, when you cannot run the experiment you wish you could.

13 published

  1. Why a great model answers the wrong question
  2. Potential outcomes and counterfactuals
  3. Confounding
  4. Causal DAGs and what to control for
  5. Colliders and selection bias
  6. Simpson's paradox
  7. Propensity score matching
  8. Inverse probability weighting
  9. Difference-in-differences
  10. Regression discontinuity
  11. Instrumental variables
  12. Uplift modelling
  13. Double machine learning

Other ways to use this site

Learning paths

An ordered route through the lessons for one job: ML developer, AI engineer, computer vision engineer.

Projects

Build real things with full code, a dataset and a deployment guide.

AI glossary

Every term you keep seeing, defined in one plain sentence first.

Error database

Paste the error you got. Find out what it means and how to fix it.