辅助任务和探测启用ObjectNav
Auxiliary Tasks and Exploration Enable ObjectNav
摘要：
ObjectNav（ObjectGoal Navigation）是一个具体的任务，其中代理将导航到一个看不见的环境中的对象实例。先前的工作已经表明，使用普通视觉和递归模块的端到端ObjectNav代理，例如CNN+RNN，由于过度拟合和样本效率低下而表现不佳。这促使目前最先进的方法混合分析和学习的组件，并在明确的空间地图的环境。相反，我们通过添加辅助学习任务和探索奖励来重新启用通用学习代理。在Habitat ObjectNav Challenge上，我们的代理实现了24.5%的成功率和8.1%的SPL，分别比现有技术水平提高了37%和8%。根据我们的分析，我们提出代理将采取行动简化其视觉输入，以平滑其RNN动态，辅助任务通过最小化有效RNN维度来减少过拟合;即，必须在长期范围内保持连贯计划的高性能ObjectNav代理通过学习平滑，低维的递归动态来实现。网址：https://joel99.github.io/objectnav/
英文摘要：
ObjectGoal Navigation (ObjectNav) is an embodied task wherein agents are to navigate to an object instance in an unseen environment. Prior works have shown that end-to-end ObjectNav agents that use vanilla visual and recurrent modules, e.g. a CNN+RNN, perform poorly due to overfitting and sample inefficiency. This has motivated current state-of-the-art methods to mix analytic and learned components and operate on explicit spatial maps of the environment. We instead re-enable a generic learned agent by adding auxiliary learning tasks and an exploration reward. Our agents achieve 24.5% success and 8.1% SPL, a 37% and 8% relative improvement over prior state-of-the-art, respectively, on the Habitat ObjectNav Challenge. From our analysis, we propose that agents will act to simplify their visual inputs so as to smooth their RNN dynamics, and that auxiliary tasks reduce overfitting by minimizing effective RNN dimensionality; i.e. a performant ObjectNav agent that must maintain coherent plans over long horizons does so by learning smooth, low-dimensional recurrent dynamics. Site: https://joel99.github.io/objectnav/
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。
