辍学的梦想之地：从学习模拟器到现实的概括
Dropout's Dream Land: Generalization from Learned Simulators to Reality
摘要：
世界模型是用于模拟环境的生成模型。世界模型已经被证明能够学习强化学习环境的空间和时间表示。在某些情况下，世界模型为智能体提供了完全在自己的梦境环境中学习的机会。在这项工作中，我们将探索如何提高从梦境环境到真实环境的泛化能力（Dream 2 Real）。我们提出了一种通用的方法来提高控制器的能力，从一个神经网络的梦想环境转移到现实中，在很少的额外成本。这些改进是通过从Domain Randomization中汲取灵感而获得的，Domain Randomization的基本思想是在不从根本上改变手头任务的情况下尽可能多地随机化模拟器。通常，域随机化假设可以访问具有可配置参数的预构建模拟器，但通常这是不可用的。通过使用dropout训练世界模型，梦境环境能够创建几乎无限数量的不同梦境环境。先前的dropout用例要么在推理时不使用dropout，要么对多个采样掩码生成的预测进行平均（Monte-Carlo Dropout）。Dropout's Dream Land利用每个独特的面具来创造一系列不同的梦境环境。我们的实验结果表明，辍学的梦想土地是一个有效的技术，以弥合现实之间的差距，梦想环境和现实。此外，我们还进行了一系列广泛的消融研究。
英文摘要：
A World Model is a generative model used to simulate an environment. World Models have proven capable of learning spatial and temporal representations of Reinforcement Learning environments. In some cases, a World Model offers an agent the opportunity to learn entirely inside of its own dream environment. In this work we explore improving the generalization capabilities from dream environments to real environments (Dream2Real). We present a general approach to improve a controller's ability to transfer from a neural network dream environment to reality at little additional cost. These improvements are gained by drawing on inspiration from Domain Randomization, where the basic idea is to randomize as much of a simulator as possible without fundamentally changing the task at hand. Generally, Domain Randomization assumes access to a pre-built simulator with configurable parameters but oftentimes this is not available. By training the World Model using dropout, the dream environment is capable of creating a nearly infinite number of different dream environments. Previous use cases of dropout either do not use dropout at inference time or averages the predictions generated by multiple sampled masks (Monte-Carlo Dropout). Dropout's Dream Land leverages each unique mask to create a diverse set of dream environments. Our experimental results show that Dropout's Dream Land is an effective technique to bridge the reality gap between dream environments and reality. Furthermore, we additionally perform an extensive set of ablation studies.
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。
