A structured L-BFGS method and its application to inverse problems
Many inverse problems are phrased as optimization problems in which the objective
function is the sum of a data-fidelity term and a regularization. Often, the Hessian of the …
function is the sum of a data-fidelity term and a regularization. Often, the Hessian of the …
A globalization of L-BFGS for nonconvex unconstrained optimization
F Mannel - arxiv preprint arxiv:2401.03805, 2024 - arxiv.org
We present a modification of the limited memory BFGS (L-BFGS) method that ensures global
and linear convergence on nonconvex objective functions. Importantly, the modified method …
and linear convergence on nonconvex objective functions. Importantly, the modified method …
A structured L-BFGS method with diagonal scaling and its application to image registration
F Mannel, HO Aggrawal - Journal of Mathematical Imaging and Vision, 2025 - Springer
We devise an L-BFGS method for optimization problems in which the objective is the sum of
two functions, where the Hessian of the first function is computationally unavailable while the …
two functions, where the Hessian of the first function is computationally unavailable while the …
Useful Compact Representations for Data-Fitting
JJ Brust - arxiv preprint arxiv:2403.12206, 2024 - arxiv.org
For minimization problems without 2nd derivative information, methods that estimate
Hessian matrices can be very effective. However, conventional techniques generate dense …
Hessian matrices can be very effective. However, conventional techniques generate dense …