Bootstrapping a DQN Replay Memory with Synthetic Experiences
This addresses a bottleneck in DQN training for reinforcement learning agents, though it appears incremental as it builds on existing replay memory methods.
The paper tackles the problem of slow learning in Deep Reinforcement Learning by introducing an Interpolated Experience Replay algorithm that creates synthetic experiences in nondeterministic discrete environments, showing it helps the agent learn faster and better than the classic version on the FrozenLake environment.
An important component of many Deep Reinforcement Learning algorithms is the Experience Replay which serves as a storage mechanism or memory of made experiences. These experiences are used for training and help the agent to stably find the perfect trajectory through the problem space. The classic Experience Replay however makes only use of the experiences it actually made, but the stored samples bear great potential in form of knowledge about the problem that can be extracted. We present an algorithm that creates synthetic experiences in a nondeterministic discrete environment to assist the learner. The Interpolated Experience Replay is evaluated on the FrozenLake environment and we show that it can support the agent to learn faster and even better than the classic version.