Faster Deep Reinforcement Learning with Slower Online Network
This work addresses robustness issues in deep reinforcement learning for agents, though it appears incremental as it builds on existing methods with a simple modification.
The paper tackled the problem of noisy updates in deep reinforcement learning by incentivizing the online network to stay close to the target network, resulting in significant performance improvements on the Atari benchmark for DQN and Rainbow algorithms.
Deep reinforcement learning algorithms often use two networks for value function optimization: an online network, and a target network that tracks the online network with some delay. Using two separate networks enables the agent to hedge against issues that arise when performing bootstrapping. In this paper we endow two popular deep reinforcement learning algorithms, namely DQN and Rainbow, with updates that incentivize the online network to remain in the proximity of the target network. This improves the robustness of deep reinforcement learning in presence of noisy updates. The resultant agents, called DQN Pro and Rainbow Pro, exhibit significant performance improvements over their original counterparts on the Atari benchmark demonstrating the effectiveness of this simple idea in deep reinforcement learning. The code for our paper is available here: Github.com/amazon-research/fast-rl-with-slow-updates.