Qualitative similarities and differences in visual object representations between brains and deep networks.
Qualitative similarities and differences in visual object representations between brains and deep networks.
复制标题
DOI：
10.1038/s41467-021-22078-3
复制
发表时间：
2021-03-25
影响因子：
16.6
通讯作者：
Arun SP
中科院分区：
文献类型：
--
作者：
Jacob G;Pramod RT;Katti H;Arun SP
深层的神经网络彻底改变了计算机的视觉，它们的对象表示与大脑中的视觉皮质区域保持一致。在距离比较方面，神经现象在远前置的深层神经网络中询问接受对象识别的深度神经网络。在随机初始化的网络中存在某些现象，例如在物体识别训练之后的全球优势效应，稀疏性和相对大小。相关的稀疏性。这些发现表明了这些现象在大脑和深层网络中出现的足够条件，并为可以纳入的属性提供了线索，以改善深层网络。
深度神经网络在这里被广泛认为是生物视觉的良好模型，我们描述了大脑和深层网络之间的几种定性相似性和差异，这些相似之处是何时可以将深网络视为生物愿景的良好模型以及如何改善它们。
Deep neural networks have revolutionized computer vision, and their object representations across layers match coarsely with visual cortical areas in the brain. However, whether these representations exhibit qualitative patterns seen in human perception or brain representations remains unresolved. Here, we recast well-known perceptual and neural phenomena in terms of distance comparisons, and ask whether they are present in feedforward deep neural networks trained for object recognition. Some phenomena were present in randomly initialized networks, such as the global advantage effect, sparseness, and relative size. Many others were present after object recognition training, such as the Thatcher effect, mirror confusion, Weber’s law, relative size, multiple object normalization and correlated sparseness. Yet other phenomena were absent in trained networks, such as 3D shape processing, surface invariance, occlusion, natural parts and the global advantage. These findings indicate sufficient conditions for the emergence of these phenomena in brains and deep networks, and offer clues to the properties that could be incorporated to improve deep networks. Deep neural networks are widely considered as good models for biological vision. Here, we describe several qualitative similarities and differences in object representations between brains and deep networks that elucidate when deep networks can be considered good models for biological vision and how they can be improved.
登录
查看更多内容
影响因子：
25
作者：
Kar K;Kubilius J;Schmidt K;Issa EB;DiCarlo JJ
通讯作者：
DiCarlo JJ
DOI：
10.1073/pnas.1109168108
发表时间：
2011-10-25
影响因子：
11.1
作者：
Fleuret, Francois;Li, Ting;Geman, Donald
通讯作者：
Geman, Donald
影响因子：
1.8
作者：
Arun, S. P.
通讯作者：
Arun, S. P.
影响因子：
2.6
作者：
BARTLETT, JC;SEARCY, J
通讯作者：
SEARCY, J
影响因子：
1.8
作者：
Jacob G;Arun SP
通讯作者：
Arun SP