# Errata for Thoughtful Machine Learning

The errata list is a list of errors and their corrections that were found after the product was released.

The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.

Color Key: Serious Technical Mistake Minor Technical Mistake Language or formatting error Typo Question Note Update

Version Location Description Submitted By Date Submitted
Printed Page 53
In Probability of Symbols section

The Probability of Symbols section is impossible to follow. There's not enough information. Each time, I review it appears wrong. Here are the topics that are unclear: total = 6.0 Total for what? There are 5 numbers [1,2,3,4,5] not 6. p_a_cap_b = (a & b).length / total What does p_a_cap_b mean? Where does the "cap" come in? p_a_cap_b = (a & b).length / total Does it then equal to 1/6 = .16666? p_b = b.length / total = 3/6 = .5? But this can't be right since below it says that P(B)=20%. If p_b = .2 = 3/total then total = 3/.2 = 15 Where does 15 come from? Then b.length cannot equal to 3. What does it equal then? p_a_cab_b = p_a_given_b * p_b = 0.33*.0.2 = 0.06 But 0.06 does not equal to 1/6 or 1/5. Where does 0.06 come from? If you could provide additional details, that would be awesome. I really want to understand this example. Thanks very much, -Larry

Larry Freeman  Dec 18, 2014
Printed Page 53
9th line of text from the top

For the code inside the box "Probability Symbols", there is a line "total = 6.0". I believe this should be 5, not 6. If 'total' means the number of possible outcomes, assuming that events A and B are jointly exhaustive, then there are a total of 5 possible outcomes, as shown several lines earlier with the line starting "a | b". This doesn't affect the final value of p_a_given_b, because (1/total) is a factor of both p_a_cap_b and p_b, so when the former is divided by the latter, the factors cancel out. In essence, B has three outcomes, one of them is shared with A, so the probability of A given B is 1/3. This all rests on the assumption that the five outcomes are equally probable (each has a probability of .2). I am assuming that the content of Fig. 4-1 is completely independent of the content of the box "Probability Symbols".

Bruce Tesar  May 10, 2018