Deep reinforcement learning algorithms typically require vast amounts of data to train to a useful level of performance. Each time new data is encountered, the network must inefficiently update all of its parameters. Auxiliary memory units can help deep neural networks train more efficiently by separating computation from storage, and providing a means to rapidly store and retrieve precise information. We present four deep reinforcement learning models augmented with external memory, and benchmark their performance on ten tasks from the Arcade Learning Environment. Our discussion and insights will be helpful for future RL researchers developing their own memory agents.
College and Department
Physical and Mathematical Sciences
BYU ScholarsArchive Citation
Suggs, Sterling, "Reinforcement Learning with Auxiliary Memory" (2021). Theses and Dissertations. 9028.
Reinforcement learning, auxiliary memory, neural computer, Atari, machine learning, Q-learning