TY  - RPRT
TI  - A Nearly Optimal and Agnostic Algorithm for Properly Learning a Mixture of k Gaussians, for any Constant k
AU  - Jerry Li
AU  - Ludwig Schmidt
PY  - 2015
UR  - https://arxiv.org/abs/1506.01367
ID  - 1506.01367
ER  -