Averaged-DQN: Variance Reduction and Stabilization for Deep Reinforcement Learning
This addresses stability issues for DRL practitioners, offering an incremental improvement over existing methods.
The paper tackled instability and variability in Deep Reinforcement Learning by proposing Averaged-DQN, an extension that averages previous Q-value estimates, resulting in more stable training and improved performance with significant gains on the Arcade Learning Environment benchmark.
Instability and variability of Deep Reinforcement Learning (DRL) algorithms tend to adversely affect their performance. Averaged-DQN is a simple extension to the DQN algorithm, based on averaging previously learned Q-values estimates, which leads to a more stable training procedure and improved performance by reducing approximation error variance in the target values. To understand the effect of the algorithm, we examine the source of value function estimation errors and provide an analytical comparison within a simplified model. We further present experiments on the Arcade Learning Environment benchmark that demonstrate significantly improved stability and performance due to the proposed extension.