Entropy gets introduced as a formula — −Σ p log p — and the intuition evaporates on contact. This is the refresher I wish I'd had: entropy is just average surprise, measured in yes/no questions. Drag a coin's bias, reshape a distribution, and watch a sampler's running surprise converge onto the entropy, live in your browser.

You already know Hopfield networks slide downhill into the nearest memory — and get stuck. Add a single ingredient, temperature, and that deterministic descent becomes a Boltzmann machine that samples, escapes local minima, and learns. A hands-on, visual refresher with live energy landscapes, annealing, and a trainable RBM.

What should a 3rd grader do after school — study, play, or rest? Turns out this is a Markov Decision Process, and we can solve it optimally. An interactive introduction to MDPs and value iteration.