http://arxiv.org/abs/1602.04938 I find this interesting less for its practical application and more for its algorithmic formulation of "explanation"
-
@BagelDaughter an explanation is a minimally complex model approximating a local portion of a highly complex model's input space
@BagelDaughter can this technique be used to 1) automate more of the learning process? (i.e. training set selection)
-
-
@BagelDaughter 2) teach humans things originally learned by black-box learners? (i.e. how to be even better at Go)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.