For active learning, can we apply some optimisation method instead of evaluate all candidates? - Githubissues

dongwookim-ml / almc

active learning algorithms for matrix and tensor completion

GNU General Public License v2.0

8 stars 2 forks source link

For active learning, can we apply some optimisation method instead of evaluate all candidates? #2

Open dongwookim-ml opened 8 years ago

dongwookim-ml commented 8 years ago

Approximated argmax?