Emergent Neural Network Mechanisms for Generalization to Objects in Novel Orientations
Computer Vision and Pattern Recognition
2023-07-14 v2 Artificial Intelligence
Machine Learning
Neurons and Cognition
Machine Learning
Abstract
The capability of Deep Neural Networks (DNNs) to recognize objects in orientations outside the distribution of the training data is not well understood. We present evidence that DNNs are capable of generalizing to objects in novel orientations by disseminating orientation-invariance obtained from familiar objects seen from many viewpoints. This capability strengthens when training the DNN with an increasing number of familiar objects, but only in orientations that involve 2D rotations of familiar orientations. We show that this dissemination is achieved via neurons tuned to common features between familiar and unfamiliar objects. These results implicate brain-like neural mechanisms for generalization.
Cite
@article{arxiv.2109.13445,
title = {Emergent Neural Network Mechanisms for Generalization to Objects in Novel Orientations},
author = {Avi Cooper and Xavier Boix and Daniel Harari and Spandan Madan and Hanspeter Pfister and Tomotake Sasaki and Pawan Sinha},
journal= {arXiv preprint arXiv:2109.13445},
year = {2023}
}