Accession Number:

AD1090441

Title:

Harnessing Parameterization For Fast And Reliable Nonconvex Optimization

Descriptive Note:

Technical Report,05 Apr 2018,04 Oct 2019

Corporate Author:

California Univ Berkeley Berkeley United States

Report Date:

2019-10-04

Pagination or Media Count:

15.0

Abstract:

This project focused on developing novel understanding of large-scale, non-convex optimization problems by establishing robust notions of how the choice of parameterization affects the geometric and computational character of the optimization process. This understanding was used to create a methodological link between machine learning and optimal control, enabling a car to be successfully taught to drive around an unspecified track using vision-based control. Reparameterization also provided benefits for optimization of recurrent neural networks. Insights were gained into the construction of well-performing stable recurrent models for future used in machine learning.

Subject Categories:

  • Cybernetics
  • Operations Research

Distribution Statement:

APPROVED FOR PUBLIC RELEASE