Harnessing Parameterization For Fast And Reliable Nonconvex Optimization
Technical Report,05 Apr 2018,04 Oct 2019
California Univ Berkeley Berkeley United States
Pagination or Media Count:
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.
- Operations Research