Non-stationary Signal Analysis with Applications to Blind Source Processing and Imaging

Abstract

The main objective of our research program is to develop new mathematical theories, along with effective methods, efficient algorithms, and near real-time implementation schemes, for the decomposition of any(blind--source) real-world signal or time series into its primary signal building blocks, called atomic components; thereby introducing innovative powerful signal processing methods for time--frequency analysis of non--stationary and non--linear signals. The mathematical tools developed in our work can also be applied to the study of machine learning problems, and in particular, to gain some valuable insight in the expressive power of deep neural networks.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Oct 02, 2018
Accession Number
AD1068399

Entities

People

  • Hrushikesh Mhaskar

Organizations

  • Claremont Graduate University

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Applied Mathematics
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Artificial Neural Networks
  • Big Data
  • Computational Science
  • Computations
  • Computer Languages
  • Computer Science
  • Data Analysis
  • Data Science
  • Data Set
  • Deep Learning
  • Differential Equations
  • Digital Data
  • Dimensionality Reduction
  • Frequency
  • Harmonic Analysis
  • Machine Learning
  • Mathematical Analysis
  • Mathematical Models
  • Mathematics
  • Natural Languages
  • Neural Networks
  • Statistics

Fields of Study

  • Engineering

Readers

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

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks