2 ms·
My paper gives guarantees for any smooth convex minimization problem. The semidefinite programming is needed to solve a meta-optimization problem in the analysi
by ProfGrimmer 3y ago
My paper gives guarantees for any smooth convex minimization problem. The semidefinite programming is needed to solve a meta-optimization problem in the analysis of these generic problems. It's a tool rather than the target.
- bainsfather 3y agoI read the Quanta article because I am interested in optimisations for deep neural nets. Am I right in concluding that your work doesn't apply to that area, as it is looking at a much 'easier' case?