论文标题
学习何时何地缩小深度加固学习
Learning When and Where to Zoom with Deep Reinforcement Learning
论文作者
论文摘要
尽管高分辨率图像比其较低分辨率对应物包含的语义上更有用的信息在计算上更昂贵，并且在某些应用中，例如遥感，它们的获取可能要贵得多。由于这些原因，希望开发一种自动方法，在必要时选择性地使用高分辨率数据，同时保持准确性并降低获取/运行时成本。在这个方向上，我们提出了PatchDrop A的加固学习方法，以动态地识别在配对，廉价，低分辨率图像的何时何地使用/获取高分辨率数据。我们在CIFAR10，CIFAR100，ImageNet和FMOW数据集上进行实验，在该数据集中，我们使用明显较小的分辨率数据，同时保持与使用全高分辨率图像的模型相似的精度。
While high resolution images contain semantically more useful information than their lower resolution counterparts, processing them is computationally more expensive, and in some applications, e.g. remote sensing, they can be much more expensive to acquire. For these reasons, it is desirable to develop an automatic method to selectively use high resolution data when necessary while maintaining accuracy and reducing acquisition/run-time cost. In this direction, we propose PatchDrop a reinforcement learning approach to dynamically identify when and where to use/acquire high resolution data conditioned on the paired, cheap, low resolution images. We conduct experiments on CIFAR10, CIFAR100, ImageNet and fMoW datasets where we use significantly less high resolution data while maintaining similar accuracy to models which use full high resolution images.
