Next: Question 5
Up: Evaluating Hypothesis and Experimental
Previous: Comparing Learning Algorithms
- did not measure variability due to choice of
training set or internal randomness - only use if we believe these to
be small
- it does not directly compare performance on training
sets of size but only - we must assume the relative
difference will still hold for training sets of size
Patricia Jean Riddle
Wed Jun 23 13:06:34 NZST 1999