Focusing on the Big Picture: Insights into an End-to-End Systems Approach to Deep Learning for Satellite Imagery

Abstract

Deep learning tasks are often complicated and require a variety of components working together efficiently to perform well. Due to the often large scale of these tasks, there is a necessity to iterate quickly in order to attempt a variety of methods and to find and fix bugs. While participating in IARPAs Functional Map of the World challenge, we identified challenges along the entire deep learning pipeline and found various solutions to these challenges. In this paper, we present the performance, engineering, and deep learning considerations with obtaining, processing, and modeling data, as well as underlying infrastructure considerations that support large-scale deep learning tasks. We also discuss insights and observations with regard to satellite imagery and deep learning for image classification.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2019
Accession Number
AD1090851

Entities

People

  • Carson D. Sestili
  • Javier A. Vazquez-trejo
  • Matthew E. Gaston
  • Ritwik Gupta

Organizations

  • Carnegie Mellon University

Tags

DTIC Thesaurus Topics

  • Artificial Intelligence Software
  • Aspect Ratio
  • Computing System Architectures
  • Convolutional Neural Networks
  • Data Mining
  • Data Processing
  • Data Science
  • Deep Learning
  • Dimensionality Reduction
  • Engineering
  • Image Processing
  • Information Science
  • Machine Learning
  • Network Architecture
  • Neural Networks
  • Satellite Imaging
  • Software Development

Fields of Study

  • Computer science

Readers

  • Distributed Systems and Data Platform Development
  • Educational Psychology
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Space
  • Space - Space Objects