ViZDoom: DRQN with Prioritized Experience Replay, Double-Q Learning, & Snapshot Ensembling
This is an incremental improvement for reinforcement learning in complex, partially observable game environments like ViZDoom.
The paper tackled improving agent performance in the ViZDoom environment by testing double-Q learning and prioritized experience replay with a DRQN architecture, and found that both variants exceeded built-in AI agents, with snapshot ensembling further enhancing performance.
ViZDoom is a robust, first-person shooter reinforcement learning environment, characterized by a significant degree of latent state information. In this paper, double-Q learning and prioritized experience replay methods are tested under a certain ViZDoom combat scenario using a competitive deep recurrent Q-network (DRQN) architecture. In addition, an ensembling technique known as snapshot ensembling is employed using a specific annealed learning rate to observe differences in ensembling efficacy under these two methods. Annealed learning rates are important in general to the training of deep neural network models, as they shake up the status-quo and counter a model's tending towards local optima. While both variants show performance exceeding those of built-in AI agents of the game, the known stabilizing effects of double-Q learning are illustrated, and priority experience replay is again validated in its usefulness by showing immediate results early on in agent development, with the caveat that value overestimation is accelerated in this case. In addition, some unique behaviors are observed to develop for priority experience replay (PER) and double-Q (DDQ) variants, and snapshot ensembling of both PER and DDQ proves a valuable method for improving performance of the ViZDoom Marine.