TY  - RPRT
TI  - Variance-reduced $Q$-learning is minimax optimal
AU  - Martin J. Wainwright
PY  - 2019
UR  - https://arxiv.org/abs/1906.04697
ID  - 1906.04697
ER  -