Year: 2015
Type: preprint
Abstract: Training deep feature hierarchies to solve supervised learning tasks has achieved state of the art performance on many problems in computer vision. However, a principled way in which to train such hierarchies in the unsupervised setting has remained elusive. In this work we suggest a new architecture and loss for training deep feature hierarchies that linearize the transformations observed in unlabeled natural video sequences. This is done by training a generative model to predict video frames. We also address the problem of inherent uncertainty in prediction by introducing latent variables that are non-deterministic functions of the input into the network architecture.
Source: arXiv (Cornell University)
Authors: Goroshin, Ross, Mathieu, Michael, LeCun, Yann
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Funder: Nvidia
Awards: G1833903074, G2010355708, G5801408398
OpenAlex ID: W1782879880