dc.contributor.author | Xiong, Xin | |
dc.date.accessioned | 2014-01-23 21:14:25 (GMT) | |
dc.date.available | 2014-01-23 21:14:25 (GMT) | |
dc.date.issued | 2014-01-23 | |
dc.date.submitted | 2014 | |
dc.identifier.uri | http://hdl.handle.net/10012/8197 | |
dc.description.abstract | This thesis is concerned with the efficient computation of Jacobian matrices of nonlinear
vector maps using automatic differentiation (AD). Specifically, we propose the use of
two directed edge separator methods, the weighted minimum separator and natural order separator methods, to exploit the structure of the computational graph of the nonlinear system.This allows for the efficient determination of the Jacobian matrix using AD software. We will illustrate the promise of this approach with computational experiments. | en |
dc.language.iso | en | en |
dc.publisher | University of Waterloo | en |
dc.subject | Automatic differentiation | en |
dc.subject | Forward mode | en |
dc.subject | Reverse mode | en |
dc.subject | Directed acyclic graph | en |
dc.subject | Computational graph | en |
dc.subject | Directed edge separator | en |
dc.subject | Jacobian matrix | en |
dc.subject | Newton step | en |
dc.subject | Minimum cutset | en |
dc.subject | Ford-Fulkerson algorithm | en |
dc.subject | Sparsity technique | en |
dc.subject | Hidden structure | en |
dc.title | Efficient Jacobian Determination by Structure-Revealing Automatic Differentiation | en |
dc.type | Master Thesis | en |
dc.pending | false | |
dc.subject.program | Combinatorics and Optimization | en |
uws-etd.degree.department | Combinatorics and Optimization | en |
uws-etd.degree | Master of Mathematics | en |
uws.typeOfResource | Text | en |
uws.peerReviewStatus | Unreviewed | en |
uws.scholarLevel | Graduate | en |