3 ms·
It's fine to talk about statistical accuracy when the distribution is somewhat normal. ML systems we've seen so far don't typically degrade in nice, statistica
by nitrogen 8y ago
It's fine to talk about statistical accuracy when the distribution is somewhat normal. ML systems we've seen so far don't typically degrade in nice, statistical, predictable ways.
Omitting 90% of samples (based on 10x speed claim) leaves an awful lot of room for bizarre errors.