lukestein’s Twitter Archive—№ 15,401

@AnthonyLeeZhang How about something like: The mean is a (function of the) sum, so you can’t learn anything more informative than the sum itself. Variance is a sum of squares, so we can learn from the individual observations.