4 problems

0 votes0 replies0 views
Worst-case performance conjecture for gradient descent on linear-operator compositions
Let $g$ be a $mu_g$ -strongly convex and $L_g$ -smooth function, and let $M$ belong to the class of linear operators with singular values in $[mu_M,L_M]$ . Write…
Conjectureopengradient-method optimization worst-case-analysisarXiv posted 2 days ago
0 votes0 replies1 view
Closed-form worst-case conjecture for gradient descent on $\mathcal{D}_{\mu_g}^{\mu_M}$
Closed-form worst-case conjecture. The worst-case performance satisfies
Conjectureopenconvex-optimization gradient-method performance-estimationarXiv posted 2 days ago
0 votes0 replies1 view
Taylor et al.'s one-dimensional worst-case conjecture for gradient descent on smooth strongly convex functions
Let $N\geq 0$ , $h\in[0,2]$ , and let $\mathcal{F}_{\mu_f}$ be the class of functions with the smoothness and strong-convexity parameters specified in the source. Let…
Conjectureopenconvex-optimization gradient-method performance-estimationarXiv posted 2 days ago
0 votes0 replies0 views
Exact complexity bound conjecture for the gradient method
Let $f$ be a convex function with Lipschitz-continuous gradient of constant $L$ , let $x_\ast$ be a minimizer, and let $R=\|x_0-x_\ast\|$ . Suppose the sequence $x_0,\dots,x_N$ is ge…
Conjectureopenconvex-optimization first-order-methods gradient-methodarXiv posted 2 days ago

Worst-case performance conjecture for gradient descent on linear-operator compositions