01-09-2016

Unsupervised Deep learning

Deep learning is a branch of machine learning dedicated to algorithms which find abstractions or partial representations of data through multiple layers of neurons or other complex structures. Unsupervised Deep Learning studies neural architectures which learn from data without requiring labels.

Published works:

  • A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines. D. Charte, F. Charte, S. GarcĂ­a, M J. del Jesus, F. Herrera. (Information Fusion, 2018) ScienceDirect, arXiv

Developed software:

  • Ruta, an early-stage R package for training unsupervised deep learning techniques such as autoencoders
  • Rutavis, a companion package for visualizations