Building Generalizable Agents with a Realistic and Rich 3D Environment
Yi Wu UC Berkeley、Yuxin Wu、Georgia Gkioxari、Yuandong Tian Facebook AI Research {yuxinwu, gkioxari,
Abstract
Teaching an agent to navigate in an unseen 3D environment is a challenging task, even in the event of simulated environments. To generalize to unseen environments, an agent needs to be robust to low-level variations (e.g. color, texture, object changes), and also high-level variations (e.g. layout changes of the environment). To improve overall generalization, all types of variations in the environment have to be taken under consideration via different level of data augmentation steps. To this end, we propose House3D, a rich, extensible and efficient environment that contains 45,622 human-designed 3D scenes of visually realistic houses, ranging from single-room studios to multi-storied houses, equipped with a diverse set of fully labeled 3D objects, textures and scene layouts, based on the SUNCG dataset (Song et al., 2017). The diversity in House3D opens the door towards scene-level augmentation, while the label-rich nature of House3D enables us to inject pixel- & task-level augmentations such as domain randomization Tobin et al. (2017) and multi-task training. Using a subset of houses in House3D, we show that reinforcement learning agents trained with an enhancement of different l
中文速览
让智能体在没见过的3D室内环境中自主导航是强化学习的一大难题，核心挑战在于如何让智能体泛化到颜色、纹理、物体摆放乃至整体房间布局都完全不同的新场景。为此，研究者基于SUNCG数据集构建了名为House3D的大规模仿真平台，包含超过4.5万个人工设计的三维房屋场景，并在此基础上设计了"RoomNav"导航任务——让智能体根据"去厨房"这样的语义指令找到目标房间。他们同时从像素、物体和场景三个层次对训练数据进行增强（domain randomization、多概念联合训练、多样化场景），配合门控LSTM策略网络和A3C算法训练智能体，最终在50个从未见过的房屋中导航成功率达到35.8%，比仅使用原始RGB输入的基线方法高出约10个百分点。这项工作表明，要让强化学习智能体真正泛化到陌生的真实感三维场景，必须同时在多个层次上引入数据多样性，而House3D作为开源平台也为后续室内导航、具身问答等研究提供了重要基础。
原文 arXiv:1801.02209;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1801.02209v2