Maximum Entropy Dueling Network Architecture in Atari Domain
This work addresses policy evaluation challenges in reinforcement learning for Atari domains, but it is incremental as it builds upon existing Dueling Networks.
The paper tackled the problem of policy evaluation in reinforcement learning for Atari games by proposing an improved Dueling Network architecture based on Maximum Entropy, resulting in better performance compared to the original network and other value-based methods.
In recent years, there have been many deep structures for Reinforcement Learning, mainly for value function estimation and representations. These methods achieved great success in Atari 2600 domain. In this paper, we propose an improved architecture based upon Dueling Networks, in this architecture, there are two separate estimators, one approximate the state value function and the other, state advantage function. This improvement based on Maximum Entropy, shows better policy evaluation compared to the original network and other value-based architectures in Atari domain.