论文标题
用catalyst.rl样品有效的合奏学习
Sample Efficient Ensemble Learning with Catalyst.RL
论文作者
论文摘要
我们提出了Catalyst.rl，这是一个用于可再现和样品有效增强学习（RL）研究的开源Pytorch框架。 catalyst.rl的主要特征包括大规模的分布式分布式培训，各种RL算法和辅助技巧的有效实现，例如N-步骤回报，价值分布，增值，增强加固学习等。人类肌肉骨骼模型的控制器。环境在计算上很昂贵，具有高维的连续动作空间，并且是随机的。我们的团队获得了第二名，利用了Catalyst.rl在短短几个小时的培训时间内培训高质量和样品效率RL代理的能力。该实现与实验是开源的，因此可以再现结果并尝试了新颖的想法。
We present Catalyst.RL, an open-source PyTorch framework for reproducible and sample efficient reinforcement learning (RL) research. Main features of Catalyst.RL include large-scale asynchronous distributed training, efficient implementations of various RL algorithms and auxiliary tricks, such as n-step returns, value distributions, hyperbolic reinforcement learning, etc. To demonstrate the effectiveness of Catalyst.RL, we applied it to a physics-based reinforcement learning challenge "NeurIPS 2019: Learn to Move -- Walk Around" with the objective to build a locomotion controller for a human musculoskeletal model. The environment is computationally expensive, has a high-dimensional continuous action space and is stochastic. Our team took the 2nd place, capitalizing on the ability of Catalyst.RL to train high-quality and sample-efficient RL agents in only a few hours of training time. The implementation along with experiments is open-sourced so results can be reproduced and novel ideas tried out.
