Notebook

The path Wilson Pruitt is following through the primary sources of machine learning, with the notes he takes as he reads. Every source is linked and every exercise is spelled out, so anyone can take the same route. Each station links to the records it added or verified, and each record links back.

The path

Primary sources in reading order.

• Station 1
  Rosenblatt (1958), The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain not started 0 notes
• Station 2
  Minsky and Papert (1969), Perceptrons: An Introduction to Computational Geometry (introduction only) not started 0 notes
• Station 3
  Rumelhart, Hinton and Williams (1986), Learning representations by back-propagating errors not started 0 notes
• Station 4
  Vaswani et al. (2017), Attention Is All You Need not started 0 notes
• Station 5
  Radford et al. (2019), Language Models are Unsupervised Multitask Learners not started 0 notes
• Station 6
  Touvron et al. (2023), LLaMA: Open and Efficient Foundation Language Models not started 0 notes

Read alongside

Historiography: how technologies and their classifications evolve.

• Read alongside
  Basalla (1988), The Evolution of Technology not started 0 notes
• Read alongside
  Olazaran (1996), A Sociological Study of the Official History of the Perceptrons Controversy not started 0 notes
• Read alongside
  Reuleaux (1875; trans. Kennedy, 1876), The Kinematics of Machinery not started 0 notes

Exercises

Hands-on work with the records.

• Exercise
  Exercise 1: Reading a specimen, GPT-2 XL vs. Mistral 7B read 2026-09-24 0 notes