Computing & AI Class 9
How AI Learns I: From Rules to Data
Split a pile of examples and measure how often a guess is right.
11 lessons · ~80 min total
Track progress — 0 of 11 lessons read 0%
Lessons in this topic
- 0 Start here Split a pile of priced crates and count how often a guess is right. · 4 min You are here
- 1 A rule a person wrote, and a pile of examples A written rule and a pile of examples can score almost the same and still miss different crates. · 8 min
- 2 Keep some crates back Split the evening pile in two. Crates the program never saw test it fairly. · 8 min
- 3 Right out of how many Walk the guesses and the real words together. Add one for each match. · 8 min
- 4 Split it differently Move the split and the score moves too. Report the count with the size of the test set. · 8 min
- 5 Learning the examples instead of the pattern A nearest-example program copies the crates it holds. Only new crates give a fair score. · 8 min
- 6 Ask more than one Let the nearest one, three or five examples vote. More voices help on some piles and hurt on others. · 8 min
- 7 The order you do it in Collect, label, split, learn, measure. Measure on what it learned from and the score flatters. · 8 min
- 8 Write the rule yourself Walk the examples once, carry the nearest so far, and hand back its word. · 8 min
- 9 The big picture Eight ideas, one pile of crates, and one question: measured on what? · 4 min
- 10 Track quiz A cheat-sheet table and a short quiz on the eight ideas. · 8 min
0%
Take this year's practice exam →
Draws questions from every track in this subject's year, not just this one — a Practice result, never official.