One-hot encoding is a labelling system that embeds classes as standard basis vectors in a label space. Despite seeing near-universal use in supervised categorical classification tasks, the scheme is problematic in its geometric implication that, as all classes are equally distant, all classes are equally different. This is inconsistent with most, if not all, real-world tasks due to the prevalence of ancestral and convergent relationships generating a varying degree of morphological similarity across classes. We address this issue by introducing curvature to the label-space using a metric tensor as a self-regulating method that better represents these relationships as a bolt-on, learning-algorithm agnostic solution. We propose both general constraints and specific statistical parameterizations of the metric and identify a direction for future research using autoencoder-based parameterizations.
[1]
Shun-ichi Amari,et al.
Information Geometry and Its Applications
,
2016
.
[2]
Michael Bain,et al.
B-CNN: Branch Convolutional Neural Network for Hierarchical Classification
,
2017,
ArXiv.
[3]
Robinson Piramuthu,et al.
HD-CNN: Hierarchical Deep Convolutional Neural Networks for Large Scale Visual Recognition
,
2014,
2015 IEEE International Conference on Computer Vision (ICCV).
[4]
Thomas Brox,et al.
Striving for Simplicity: The All Convolutional Net
,
2014,
ICLR.
[5]
Rob Fergus,et al.
Visualizing and Understanding Convolutional Networks
,
2013,
ECCV.
[6]
Alex Krizhevsky,et al.
Learning Multiple Layers of Features from Tiny Images
,
2009
.
[7]
Priyadarshini Panda,et al.
Tree-CNN: A hierarchical Deep Convolutional Neural Network for incremental learning
,
2018,
Neural Networks.