Nonconvex optimization
What theoretical guarantees can be made about nonconvex optimization?
Blog posts:
- Rong Ge: Escaping from saddle points
- Ben Recht: Saddles again (sequel post)
- Anandkumar: NIPS 2015 workshop
Lectures:
- Ryan Tibshirani: Nonconvex?! NP! (No problem!) (slides)
Literature
- Ge et al, 2015: Escaping from saddle points: Online stochastic gradient for tensor decomposition (arxiv)
- Anandkumar & Ge, 2016: Efficient approaches for escaping higher order saddle points in non-convex optimization (arxiv)
- De Sa, Olukotun, Re, 2014: Global convergence of stochastic gradient descent for some nonconvex matrix problems (arxiv)