Emergence of brain-like mirror-symmetric viewpoint tuning in convolutional neural networks

Author:

Farzmahdi Amirhossein12ORCID,Zarco Wilbert1ORCID,Freiwald Winrich A13ORCID,Kriegeskorte Nikolaus4567ORCID,Golan Tal4ORCID

Affiliation:

1. Laboratory of Neural Systems, The Rockefeller University

2. School of Cognitive Sciences, Institute for Research in Fundamental Sciences

3. The Center for Brains, Minds & Machines

4. Zuckerman Mind Brain Behavior Institute, Columbia University

5. Department of Psychology, Columbia University

6. Department of Neuroscience, Columbia University

7. Department of Electrical Engineering, Columbia University

Abstract

Primates can recognize objects despite 3D geometric variations such as in-depth rotations. The computational mechanisms that give rise to such invariances are yet to be fully understood. A curious case of partial invariance occurs in the macaque face-patch AL and in fully connected layers of deep convolutional networks in which neurons respond similarly to mirror-symmetric views (e.g. left and right profiles). Why does this tuning develop? Here, we propose a simple learning-driven explanation for mirror-symmetric viewpoint tuning. We show that mirror-symmetric viewpoint tuning for faces emerges in the fully connected layers of convolutional deep neural networks trained on object recognition tasks, even when the training dataset does not include faces. First, using 3D objects rendered from multiple views as test stimuli, we demonstrate that mirror-symmetric viewpoint tuning in convolutional neural network models is not unique to faces: it emerges for multiple object categories with bilateral symmetry. Second, we show why this invariance emerges in the models. Learning to discriminate among bilaterally symmetric object categories induces reflection-equivariant intermediate representations. AL-like mirror-symmetric tuning is achieved when such equivariant responses are spatially pooled by downstream units with sufficiently large receptive fields. These results explain how mirror-symmetric viewpoint tuning can emerge in neural networks, providing a theory of how they might emerge in the primate brain. Our theory predicts that mirror-symmetric viewpoint tuning can emerge as a consequence of exposure to bilaterally symmetric objects beyond the category of faces, and that it can generalize beyond previously experienced object categories.

Funder

National Eye Institute

National Institute of Neurological Disorders and Stroke

Naval Research Laboratory

Charles H. Revson Foundation

Publisher

eLife Sciences Publications, Ltd

Reference73 articles.

1. Face recognition depends on specialized mechanisms tuned to view-invariant facial features: Insights from deep neural networks optimized for face or object recognition;Abudarham;Cognitive Science,2021

2. Hierarchical processing of face viewpoint in human visual cortex;Axelrod;The Journal of Neuroscience,2012

3. Why do deep convolutional networks generalize so poorly to small image transformations;Azulay;Journal of Machine Learning Research,2019

4. Face detection in untrained deep neural networks;Baek;Nature Communications,2021

5. Face;Baek,2021

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3