Recurrent Action Transformer with Memory

arXiv:2306.09459v3 Announce Type: replace
Abstract: Recently, the use of transformers in offline reinforcement learning has become a rapidly developing area. This is due to their ability to treat the agent's trajectory in the environment as a sequence, thereby reducing the policy learning problem to sequence modeling. In environments where the agent's decisions depend on past events, it is essential to capture both the event itself and the decision point in the context of the model. However, the quadratic complexity of …

abstract agent arxiv become cs.ai cs.lg decisions environment environments memory modeling offline policy reinforcement reinforcement learning the environment trajectory transformer transformers type