Multi-sample $$\zeta $$ ζ -mixup: richer, more realistic synthetic samples from a p-series interpolant

Multi-sample $$\zeta $$ ζ -mixup: richer, more realistic synthetic samples from a p-series interpolant

Abstract Modern deep learning training procedures rely on model regularization techniques such as data augmentation methods, which generate training samples that increase the diversity of data and richness of label information. A popular recent method, mixup, uses convex combinations of pairs of ori...

Full description

Bibliographic Details
Main Authors: Kumar Abhishek, Colin J. Brown, Ghassan Hamarneh
Format: Article
Language:English
Published: SpringerOpen 2024-03-01
Series:Journal of Big Data
Subjects:
Deep learning
Classification
Data augmentation
Mixup
Intrinsic dimensionality
Data manifold
Online Access:https://doi.org/10.1186/s40537-024-00898-6

Internet

https://doi.org/10.1186/s40537-024-00898-6

Similar Items