Learning Rotations
Abstract
Many different matrix classes have been tackled recently using online learning techniques, but at least one major class has been left out: rotations. We pose the online learning of rotations as an open problem and discuss the importance and challenges of this problem, including the curved manifold structure of the SO(n) rotation group which rules out direct linear combinations of rotations.
Cite
Text
Smith and Warmuth. "Learning Rotations." Annual Conference on Computational Learning Theory, 2008.Markdown
[Smith and Warmuth. "Learning Rotations." Annual Conference on Computational Learning Theory, 2008.](https://mlanthology.org/colt/2008/smith2008colt-learning/)BibTeX
@inproceedings{smith2008colt-learning,
title = {{Learning Rotations}},
author = {Smith, Adam M. and Warmuth, Manfred K.},
booktitle = {Annual Conference on Computational Learning Theory},
year = {2008},
pages = {517},
url = {https://mlanthology.org/colt/2008/smith2008colt-learning/}
}