想象后见经验回放：面向稀疏奖励任务的好奇心驱动基于模型的学习
Imaginary Hindsight Experience Replay: Curious Model-based Learning for Sparse Reward Tasks
AI总结：
本文提出想象后见经验回放方法，通过想象数据、后见经验回放和好奇心内在奖励解决稀疏奖励多目标任务，在OpenAI Gym Fetch Robotics基准上数据效率平均提升一个数量级。
摘要：
基于模型的强化学习因其相对于无模型方法更高的数据效率，成为实际机器人应用中一种有前景的学习策略。然而，当前最先进的基于模型方法依赖于成形奖励信号，而这种信号可能难以设计和实现。为解决这一问题，我们提出了一种简单的基于模型方法，专门针对稀疏奖励多目标任务，无需复杂的奖励工程。该方法被称为想象后见经验回放，通过将想象数据纳入策略更新来最小化真实世界交互。为改善稀疏奖励环境下的探索，策略使用标准后见经验回放进行训练，并赋予基于好奇心的内在奖励。在评估中，该方法在基准OpenAI Gym Fetch Robotics任务中，平均数据效率比最先进的无模型方法提高了一个数量级。
英文摘要：
Model-based reinforcement learning is a promising learning strategy for practical robotic applications due to its improved data-efficiency versus model-free counterparts. However, current state-of-the-art model-based methods rely on shaped reward signals, which can be difficult to design and implement. To remedy this, we propose a simple model-based method tailored for sparse-reward multi-goal tasks that foregoes the need for complicated reward engineering. This approach, termed Imaginary Hindsight Experience Replay, minimises real-world interactions by incorporating imaginary data into policy updates. To improve exploration in the sparse-reward setting, the policy is trained with standard Hindsight Experience Replay and endowed with curiosity-based intrinsic rewards. Upon evaluation, this approach provides an order of magnitude increase in data-efficiency on average versus the state-of-the-art model-free method in the benchmark OpenAI Gym Fetch Robotics tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
