> A neural net can approximate any continuous function.
Can't a RF or boosted regression tree theoretically do the same and then some (approximate discontinuous functions)?
Can't a RF or boosted regression tree theoretically do the same and then some (approximate discontinuous functions)?
No comments yet.