Rethinking Layer-wise Feature Amounts in Convolutional Neural Network Architectures
Abstract
We characterize convolutional neural networks with respect to the relative amount of features per layer. Using a skew normal distribution as a parametrized framework, we investigate the common assumption of monotonously increasing feature-counts with higher layers of architecture designs. Our evaluation on models with VGG-type layers on the MNIST, Fashion-MNIST and CIFAR-10 image classification benchmarks provides evidence that motivates rethinking of our common assumption: architectures that favor larger early layers seem to yield better accuracy.
- Publication:
-
arXiv e-prints
- Pub Date:
- December 2018
- DOI:
- 10.48550/arXiv.1812.05836
- arXiv:
- arXiv:1812.05836
- Bibcode:
- 2018arXiv181205836M
- Keywords:
-
- Computer Science - Machine Learning;
- Computer Science - Computer Vision and Pattern Recognition;
- Statistics - Machine Learning
- E-Print:
- Accepted at the Critiquing and Correcting Trends in Machine Learning (CRACT) Workshop at the 32nd Conference on Neural Information Processing Systems (NeurIPS 2018)