If you want to dive deeper into the question of variance and bias tradeoff in Deep Reinforcement Learning, you can check these two articles: