AIQ: How Artificial Intelligence Works and How We Can Harness Its Power For a Better World by Nick Polson and James Scott

BATB score: 7.5/10 💛

*reader discretion advised: nerd alert*

Best as: the mathematics behind GenAI; “personalization” means “conditional probability” and how Netflix customization and ratings works and the equations behind them all

Best for: Bayes’s Rule as an Equation: P (H | D) = [ P(H) + P(D|H) ] / P(D) ; where P is probability, H is Hypothesis, and D is Data

BATB lingering thought: The prevalence of breast cancer among people like Alice is 1%. That is, for every 1,000 40-year-old women who have a routine mammogram, about 10 of them have breast cancer. P(H) = 0.01. The test has an 80% detection rate: if we give it to 10 women with cancer, it will detect about 8 of those cases on average. P(D|H) = 0.8. The test has a 10% false-positive rate: if we give it to 1000 women without breast cancer, it will wrongly flag about 100 of them on average. P(D) = 0.100+0.008. Alice’s mammogram test result is positive for cancer. What is the posterior probability that it is really true positive?

Best to: 0.01*0.8 / 0.108 = 0.074 ; 7.4% chance that Alice’s positive mammogram is really true positive. Fun fact, most doctors would have answered 80% and that’s just incorrect maths (and unnecessary panic attacks).

Best summary: In AI, a “pattern” is a prediction rule that maps an input to an output. “Learning a pattern” means fitting a good prediction rule to a data set.

Best quotes: “Citizens should participate in these discussions from a position of knowledge, rather than fear, of the basic technical details. Put simply, smart people who care about the world simply must know more about AI.”

“The most dangerous phrase in the language is, ‘We’ve always done it that way’.”

“Just because something good can be done with data doesn’t mean it will be done.”

“A machine can make predictions based on the assumptions with which it’s programmed, but only people can check those assumptions.”