NASA NTRS · 19760033566
A counter-example in linear feature selection theory
Abstract
The paper shows that it is possible to construct two k x n matrices, both of which maximize divergence in the transformed space of the linear feature selection problem in multiclass pattern recognition, and which are not row equivalent. Thus, even under extremely strong conditions, it is not possible to assume that all matrix solutions which maximize transformed divergence are row equivalent.
Keep this discovery
Explore connections, maps & timelines
Brown, D. R., Omalley, M. J.. 1976-01-01. A counter-example in linear feature selection theory. https://ntrs.nasa.gov/citations/19760033566
Cite the original work for its findings. Save a collection to share your selection of sources.