2. Joke
● What's the difference between a
mathematician, an engineer and a
programmer?
3. Punchline
● Mathematicians use natural log (base e)
● Engineers use decibels (10 times log base
10)
● Programmers use bits (log base 2)
4. Useful functions
● odds(p)=p/(1-p)
– Gambler talk: 1/3 → “1-to-2”
● logit(p)=log(odds(p))
– Remember: logs are base 2, or bits
● expit(p)=exp(p/(1+p))
– Inverse of logit
5. What is belief?
● Belief(X) = logit(Probability you assign to X)
– Measured in bits
● Fun fact: Belief(not X)=-Belief(X)
6. Examples
● Belief(X)=0: probability 0.5, zero
knowledge
● Belief(X)=1: probability is 2/3
● Belief(X)=-1: probability is 1/3
● Belief(X)=5: probability about 0.97
7. More examples
● Belief(X)=10: “I’m 99.9% certain about
this!”
● Belief(X)=-10: “There’s a 0.001 chance of
that!”
● Belief(X)=infinity: probability 1, or “The
religious belief”…
8. Accuracy of belief
● Overconfidence: >>1-expit(B) of beliefs of
strength >B are wrong (for some B>0)
● Underconfidence: <<1-expit(B) of beliefs
of 0<strength<B are wrong (for some B>0)
● Well-calibrated: Neither overconfident nor
underconfident
9. Evidence
● Event E happened. Is X true?
● E is helpful only when P(E given X) != P(E
given not X). But how much?
● Likelihood(E given X) = P(E given X)/P(E
given not X)
● Evidence(E about X) = log(Likelihood(E
given X))
● Evidence is measured in bits!
10. THE FORMULA
Belief(X after seeing E) =
Belief(X)+Evidence(E about X)
11. Bayes' Theorem
● “If you are well-calibrated, and update
beliefs according to THE FORMULA, you
remain well-calibrated”
● Corrolary: If you sometimes count evidence
twice, or sometimes only weakly, you FALL
OUT OF CALIBRATION!