3 problems

Strict convergence-rate advantage of back-projection over least squares
arXiv@arxiv
Let $P_{LS}$ and $P_{BP}$ be the convergence-rate quantities associated with the least-squares and back-projection objectives, respectively, and let $cmathcal{C}_{s}(\bm{x}_{*})$ b…
Conjectureopeninverse-problems optimization projected-gradient-descent
0
Conjecture that perturbed or stochastic projected gradient descent cannot escape strict saddle points
arXiv@arxiv
Consider projected gradient descent for constrained non-convex optimization, together with versions perturbed by noise or randomized through stochastic updates. The example in the…
Conjectureopennonconvex-optimization projected-gradient-descent strict-saddles
1
Projected gradient descent convergence conjecture for matrix completion
arXiv@arxiv
Let $M^*\in\mathcal{S}^{d\times d}$ be symmetric positive semidefinite, and let $\Omega$ be generated under the model $\operatorname{SMC}(M^*,p)$ : each $(i,j)$ with $i\geq j$ is ob…
Conjectureopenmatrix-completion optimization projected-gradient-descent
0

Strict convergence-rate advantage of back-projection over least squares