TensorFlow Agents: Efficient Batched Reinforcement Learning in TensorFlow
arXiv: Learning, Volume abs/1709.02878, 2017.
We introduce TensorFlow Agents, an efficient infrastructure paradigm for building parallel reinforcement learning algorithms in TensorFlow. We simulate multiple environments in parallel, and group them to perform the neural network computation on a batch rather than individual observations. This allows the TensorFlow execution engine to p...More
Full Text (Upload PDF)
PPT (Upload PPT)