Department of Physiology, The University of Tokyo School of Medicine, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
Neural Netw. 2020 May;125:185-193. doi: 10.1016/j.neunet.2020.02.009. Epub 2020 Feb 24.
Although our brain and deep neural networks (DNNs) can perform high-level sensory-perception tasks, such as image or speech recognition, the inner mechanism of these hierarchical information-processing systems is poorly understood in both neuroscience and machine learning. Recently, Morcos et al. (2018) examined the effect of class-selective units in DNNs, i.e., units with high-level selectivity, on network generalization, concluding that hidden units that are selectively activated by specific input patterns may harm the network's performance. In this study, we revisited their hypothesis, considering units with selectivity for lower-level features, and argue that selective units are not always harmful to the network performance. Specifically, by using DNNs trained for image classification, we analyzed the orientation selectivity of individual units, a low-level selectivity widely studied in visual neuroscience. We found that orientation-selective units exist in both lower and higher layers of these DNNs, as in our brain. In particular, units in lower layers became more orientation-selective as the generalization performance improved during the course of training. Consistently, networks that generalized better were more orientation-selective in the lower layers. We finally revealed that ablating these selective units in the lower layers substantially degraded the generalization performance of the networks, at least by disrupting the shift-invariance of the higher layers. These results suggest that orientation selectivity can play a causally important role in object recognition, and that, contrary to the triviality of units with high-level selectivity, lower-layer units with selectivity for low-level features may be indispensable for generalization, at least for the several network architectures.
尽管我们的大脑和深度神经网络 (DNN) 可以执行高级的感觉感知任务，例如图像或语音识别，但在神经科学和机器学习中，这些分层信息处理系统的内部机制仍了解甚少。最近，Morcos 等人（2018 年）研究了 DNN 中类选择性单元的作用，即具有高级选择性的单元，对网络泛化的影响，得出结论认为，对特定输入模式具有选择性激活的隐藏单元可能会损害网络的性能。在这项研究中，我们重新审视了他们的假设，考虑了对较低层次特征具有选择性的单元，并认为选择性单元并不总是对网络性能有害。具体来说，我们使用经过图像分类训练的 DNN 分析了单个单元的方向选择性，这是视觉神经科学中广泛研究的一种低层次选择性。我们发现，这些 DNN 中存在于较低和较高层的具有方向选择性的单元，与我们的大脑中的情况相同。特别是，在训练过程中，随着泛化性能的提高，较低层的单元变得更加具有方向选择性。一致地，泛化性能更好的网络在较低层的方向选择性更高。我们最终揭示，在较低层中消除这些选择性单元会大大降低网络的泛化性能，至少会破坏较高层的不变性。这些结果表明，方向选择性可以在对象识别中起因果重要作用，并且与高层选择性单元的琐碎性相反，具有低级特征选择性的较低层单元对于泛化可能是必不可少的，至少对于几种网络架构而言。
