Home

Resources

• Syllabus
• Slack
• Partial timeline

Lecture Notes

(see Syllabus for overview of course)

• Lecture 1, Jan 17: Course introduction Slides
• Lecture 2, Jan 22: boyd, Wallach (setting the stakes)
• Lecture 3, Jan 29: data: it's made of people (Wallach 2014; boyd+Crawford 2012; Tufecki "Engineering" (2014))
• Lecture 4, Feb 5: societal risk, social physics (Desrosières Ch 3, Quetelet (19th c), Porter Ch 2, Gigerenzer 1.6)
• Lecture 5, Feb 12: "vulgar statistics" and "'scientific' racism" (Desrosières Ch 4, Galton, Gould Ch 3)
• Lecture 6, Feb 19: IQ, regression-v-causation (Gould on Spearman; Spearman; Freedman on Yule (19th /20th c))
• Lecture 7, Feb 26: high church mathematical statistics/hypothesis testing (Fisher, Neyman, Pearson; Gerd (20th c))
• Lecture 8, Mar 5: WWII as birth of computing w/data: Bayes, Bletchley, "Women at the Dawn"... (WWII)
• Lecture 9, Mar 19: birth+death of AI: Turing '50, Dartmouth '55, Lighthill '73 (post-war)
• Lecture 10, Mar 26: Data at "The Labs": Tukey and EDA; Chambers
• Lecture 11, April 2: Machine Learning as a Trading Zone, Breiman's "two cultures", AI2.0
• Lecture 12, April 9: 50 years of Data Science (and data engineering)
• Lecture 13, April 16: Ethics: defining & enforcing; research & industry
• Lecture 14, April 23: Persuasion architectures; surveillance capitalism; you are the product
• Lecture 15, April 30: what did we learn? (Last day of class)

Lab Notebooks (Jupyter)

• Lab 1: introduction to the notebook, data provenance
• Lab 2: Exploratory Data Analysis; Slides
• Lab 3
  • Lab 3, Part 1: Quetelet
  • Lab 3, Part 2: EDA continued
• Lab 4
  • Lab 4, Part 1: Life expectancy
    • Life expectancy data
  • Lab 4, Part 2: Simulating Galton
• Lab 5
  • Lab 5, Part 1: Yule, multiple regression and social thought
  • Lab 5, Part 2: PCA
• Lab 6: Gould + IQ, T-tests, p-value hacking
• Lab 7: Bayesian Cryptography
• Lab 8
  • Lab 8, part 1: Eliza and Expert Systems
  • Lab 8, part 2: The birth, death, and rebirth of the Perceptron
• Lab 9 not in 2018: MCMC vs models, perceptron and supervised-v-unsupervised learning
• Lab 10
  • Lab 10, part 1 Machine learning: predictive/supervised, prescriptive/reinforcement, descriptive/unsupervised (notebook authored by Su Hang)
  • Slides on "interpretability" and the history of forests and trees
  • Lab 10, part 2: forests and trees
• Lab 11 Databases and recommendation engines
• Lab 12
  • Lab 12, part 1: Privacy
  • Lab 12, part 2: Fairness, accountability, and transparency
• Lab 13 readings on autonomy and identity