Estimator Variance in Reinforcement Learning: Theoretical Problems and Practical Solutions

By Mark D. Pendrith, Malcolm R. K. Ryan. 1997

Subfields: Reinforcement Learning, Statistical Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

1. Roy Fox
2. Ari Pakman
3. Naftali Tishby
4. Sergey Levine
5. Pieter Abbeel
6. Lihong Li
7. Dale Schuurmans
8. Ofir Nachum

In Statistical Learning Theory

1. Roy Fox
2. Naftali Tishby
3. Ari Pakman

Citations in the corpus

• Addressing Function Approximation Error in Actor-Critic Methods (2018, neutral): “Function approximation error and its effect on bias and variance in reinforcement learning algorithms have been studied in prior works (Pendrith et al., 1997; Mannor et al., 2007).”
• Taming the Noise in Reinforcement Learning via Soft Updates (2015, background): “For other works that study the effect of noise in Q-learning, although without identifying the bias (7), see [26, 27, 28].”