论文标题
卷：视觉自我监督的强化学习与对象推理
ROLL: Visual Self-Supervised Reinforcement Learning with Object Reasoning
论文作者
论文摘要
当前基于图像的增强学习（RL）算法通常在整个图像上运行，而无需执行对象级别的推理。这会导致目标采样效率低下和无效的奖励功能。在本文中，我们通过结合对象级别的推理和遮挡推理来改善以前的视觉自我监督的RL。具体而言，我们使用未知的对象细分来忽略场景中的干扰因素，以更好地奖励计算和目标。我们通过采用新颖的辅助损失和训练计划来进一步实现遮挡推理。我们证明，与几个模拟的视觉控制任务中的先前方法相比，我们提出的算法，滚动学习（具有对象级别学习的增强学习）可以更快地学习，并取得更好的最终性能。项目视频和代码可从https://sites.google.com/andrew.cmu.edu/roll获得。
Current image-based reinforcement learning (RL) algorithms typically operate on the whole image without performing object-level reasoning. This leads to inefficient goal sampling and ineffective reward functions. In this paper, we improve upon previous visual self-supervised RL by incorporating object-level reasoning and occlusion reasoning. Specifically, we use unknown object segmentation to ignore distractors in the scene for better reward computation and goal generation; we further enable occlusion reasoning by employing a novel auxiliary loss and training scheme. We demonstrate that our proposed algorithm, ROLL (Reinforcement learning with Object Level Learning), learns dramatically faster and achieves better final performance compared with previous methods in several simulated visual control tasks. Project video and code are available at https://sites.google.com/andrew.cmu.edu/roll.
