BriefGPT.xyz
Nov, 2017
基于互动回放的机器人导航单次强化学习
One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay
HTML
PDF
Jake Bruce, Niko Suenderhauf, Piotr Mirowski, Raia Hadsell, Michael Milford
TL;DR
通过基于单次环境遍历构建交互世界模型、预训练视觉特征编码器和随机环境增强的方法，提出了一种在移动机器人上学习导航到固定目标并在已知环境中进行零样本迁移的方法。该方法成功应用于真实环境变化下的导航问题，避免了过拟合问题。
Abstract
Recently,
model-free reinforcement learning
algorithms have been shown to solve challenging problems by learning from extensive interaction with the environment. A significant issue with transferring this success to the
→