Small steps and giant leaps: Minimal Newton solvers for Deep Learning.

João F. Henriques Sébastien Ehrhardt Samuel Albanie Andrea Vedaldi

Published in: CoRR (2018)

Keyphrases

deep learning
unsupervised learning
unsupervised feature learning
small number
mental models
restricted boltzmann machine
machine learning
feature space
deep belief networks
deep architectures
information retrieval
training data
pattern recognition
active learning
co occurrence
dimensionality reduction