Wagatsuma Nobuhiko, Hidaka Akinori, Tamura Hiroshi
Department of Information Science, Faculty of Science, Toho University, Funabashi, Japan.
School of Science and Engineering, Tokyo Denki University, Hatoyama-machi, Japan.
Front Comput Neurosci. 2022 Sep 30;16:979258. doi: 10.3389/fncom.2022.979258. eCollection 2022.
Analysis and understanding of trained deep neural networks (DNNs) can deepen our understanding of the visual mechanisms involved in primate visual perception. However, due to the limited availability of neural activity data recorded from various cortical areas, the correspondence between the characteristics of artificial and biological neural responses for visually recognizing objects remains unclear at the layer level of DNNs. In the current study, we investigated the relationships between the artificial representations in each layer of a trained AlexNet model (based on a DNN) for object classification and the neural representations in various levels of visual cortices such as the primary visual (V1), intermediate visual (V4), and inferior temporal cortices. Furthermore, we analyzed the profiles of the artificial representations at a single channel level for each layer of the AlexNet model. We found that the artificial representations in the lower-level layers of the trained AlexNet model were strongly correlated with the neural representation in V1, whereas the responses of model neurons in layers at the intermediate and higher-intermediate levels of the trained object classification model exhibited characteristics similar to those of neural activity in V4 neurons. These results suggest that the trained AlexNet model may gradually establish artificial representations for object classification through the hierarchy of its network, in a similar manner to the neural mechanisms by which afferent transmission beginning in the low-level features gradually establishes object recognition as signals progress through the hierarchy of the ventral visual pathway.
对训练好的深度神经网络（DNN）进行分析和理解，能够加深我们对灵长类动物视觉感知中所涉及视觉机制的认识。然而，由于从各个皮层区域记录到的神经活动数据有限，在DNN的层水平上，人工神经网络与生物神经网络在视觉识别物体时的响应特征之间的对应关系仍不明确。在本研究中，我们调查了用于物体分类的训练好的AlexNet模型（基于DNN）各层中的人工表征与初级视觉皮层（V1）、中级视觉皮层（V4）和颞下皮层等不同视觉皮层水平的神经表征之间的关系。此外，我们还分析了AlexNet模型各层在单通道水平上的人工表征概况。我们发现，训练好的AlexNet模型较低层的人工表征与V1中的神经表征密切相关，而在训练好的物体分类模型的中级和中高级层中，模型神经元的反应表现出与V4神经元神经活动相似的特征。这些结果表明，训练好的AlexNet模型可能通过其网络层次逐渐建立用于物体分类的人工表征，这与传入信号从低级特征开始逐渐通过腹侧视觉通路的层次建立物体识别的神经机制类似。
