Tag: explainer

All the articles with the tag "explainer".

• Backpropagation as blame assignment

  Published:Jul 11, 2026 |  at 09:10 PM

  Backpropagation is the chain rule used as a blame ledger. I work one tiny network by hand, then scale it to the 784-16-16-10 digit network.

• What a neural network actually is, built up from one neuron

  Published:Jun 27, 2026 |  at 08:45 PM

  Starting from a single neuron and a handwritten digit, I build up what a neural network is, why it needs nonlinearity, and what training does.

• World models, from Ha and Schmidhuber to today

  Published:Jun 6, 2026 |  at 09:00 PM

  A short history of world models in AI, from a car-racing agent that learned in its own dream to Dreamer, Sora, Genie, and the Othello-GPT debate.

• Error correction: why one logical qubit needs a thousand physical ones

  Published:May 16, 2026 |  at 09:25 PM

  Why qubits fall apart, how error correction works without copying, what "below threshold" means, and what Google's Willow result actually showed.