Re:A theory of the learnable
by yters on Wednesday January 31, 2018 @01:39PM
(#16276290)
Isn't Solomonoff induction optimal if the target is learnable?
The connection to Kolmogorov complexity is interesting. It implies a perfect learner must have infinite KC or increase KC, neither of which an algorithm can do.
The connection to Kolmogorov complexity is interesting. It implies a perfect learner must have infinite KC or increase KC, neither of which an algorithm can do.