Sticky Reasoning within Learning Representations
Abstract
LeCun proposes to investigate and formalize a deep learning ("deep learning" is a new area of research in the machine learning literature) approach that recognizes various visual images (objects in images) with measureable results providing significant reduction in costs to learn and recognize images. His approach is based on Deep Convolutional Networks (ConvNets). Traditional ConvNet architectures include pooling and sub-sampling layers which reduce computational requirements, introduce invariance and prevent over-training. These benefits of pooling come at the cost of reduced localization accuracy. LeCun s research is focused on new approaches to correct the accuracy of recognition in local regions of an image.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 23, 2016
- Source ID
- FA95501510441
Entities
People
- Yann Le Cun
Organizations
- Air Force Office of Scientific Research
- New York University
- United States Air Force