On Orthogonal Projections for Dimension Reduction and Applications in Augmented Target Loss Functions for Learning Problems
Loading...
Date
2020-08-23
Authors
ORCID
Advisor
Referee
Mark
Journal Title
Journal ISSN
Volume Title
Publisher
Springer
Altmetrics
Abstract
The use of orthogonal projections on high-dimensional input and target data in learning frameworks is studied. First, we investigate the relations between two standard objectives in dimension reduction, preservation of variance and of pairwise relative distances. Investigations of their asymptotic correlation as well as numerical experiments show that a projection does usually not satisfy both objectives at once. In a standard classification problem we determine projections on the input data that balance the objectives and compare subsequent results. Next, we extend our application of orthogonal projections to deep learning tasks and introduce a general framework of augmented target loss functions. These loss functions integrate additional information via transformations and projections of the target data. In two supervised learning problems, clinical image segmentation and music information classification, the application of our proposed augmented target loss functions increase the accuracy.
Description
Citation
Journal of Mathematical Imaging and Vision. 2020, vol. 62, issue 3, p. 376-394.
https://link.springer.com/article/10.1007/s10851-019-00902-2
https://link.springer.com/article/10.1007/s10851-019-00902-2
Document type
Peer-reviewed
Document version
Published version
Date of access to the full text
Language of document
en