Mean vs mean

[Equations in this post may not look right (or appear at all) in your RSS reader. Go to the original article to see them rendered properly.] I decided to start learning about machine learning a couple of weeks ago. Partly because it sounds interesting, partly because I think I have enough math to handle it, and partly because I felt it was important to have a deeper knowledge if I want to criticize AI. I started with a library copy of Friedman and Mendhekar’s The Little Learner but decided to drop it after I was about halfway through. Although I was following along reasonably well, I kept struggling with the book’s notation—after 200 pages, I shouldn’t have to keep returning to the front of the book to see what things like t|i mean. If the book’s Scheme code were actually written in Scheme, I might have stuck with it, but its mixture of Scheme with math was too frustrating. Yesterday I started Kneusel’s Practical Deep Learning. My library gives me free access to O’Reilly ebooks, so…