Abstract
Progress in deep learning has spawned great successes in many engineering applications. As a prime example, convolutional neural networks, a type of feedforward neural networks, are now approaching -- and sometimes even surpassing -- human accuracy on a variety of visual recognition tasks. Here, however, we show that these neural networks and their recent extensions struggle in recognition tasks where co-dependent visual features must be detected over long spatial ranges. We introduce the horizontal gated-recurrent unit (hGRU) to learn intrinsic horizontal connections -- both within and across feature columns. We demonstrate that a single hGRU layer matches or outperforms all tested feedforward hierarchical baselines including state-of-the-art architectures which have orders of magnitude more free parameters. We further discuss the biological plausibility of the hGRU in comparison to anatomical data from the visual cortex as well as human behavioral data on a classic contour detection task.
Abstract (translated)
深度学习的进展在许多工程应用中取得了巨大的成功。作为一个主要的例子，卷积神经网络（一种前馈神经网络）正在接近甚至超越人类在各种视觉识别任务中的精确度。然而，在这里，我们表明这些神经网络及其最近的扩展在识别任务中挣扎，其中必须在长的空间范围内检测到相关的视觉特征。我们引入水平门控循环单元（hGRU）来学习内部水平连接 - 在特征列内和跨特征列。我们证明单个hGRU层匹配或超过所有测试过的前馈层次基线，包括具有更多数量级自由参数的最新架构。我们进一步讨论hGRU的生物可行性与来自视觉皮层的解剖数据以及经典轮廓检测任务中的人类行为数据的对比。
URL
https://arxiv.org/abs/1805.08315
