
RC
Regina Carla Sousa, Denis Vargas, Elizabeth Wanner, Joshua Knowles
· 1 min read
ResearcharXiv cs.LG
AG-BFGS: Foundations for Accelerated Geometry Learning
arXiv:2611.24113v1 Announce Type: cross
Abstract: We introduce an accelerated geometry-learning framework for BFGS that incorporates momentum into the learning of Hessian inverse approximations. We prove that the resulting accelerated updates satisfy explicit stability conditions and preserve positive definiteness within admissible parameter regions. Numerical experiments demonstrate consistent improvements over the corresponding non-accelerated schemes while maintaining numerical stability. These results provide the first rigorous theoretical foundation for accelerated quasi-Newton learning and bridge the gap between empirical observations and mathematical guarantees for accelerated second-order methods.
Original source
This story was published by arXiv cs.LG and written by Regina Carla Sousa, Denis Vargas, Elizabeth Wanner, Joshua Knowles. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


