# Retrace(λ)

> [Safe and Efficient Off-Policy Reinforcement Learning](https://arxiv.org/abs/1606.02647)