Sharper Convergence Rates for Nonconvex Optimisation via Reduction Mappings
NeurIPSSpotlight2025
TL;DR
Many high-dimensional optimisation problems exhibit rich geometric structures in their set of minimisers, often forming smooth manifolds due to over-parametrisation or symmetries…
Opening excerpt from the authors’ abstract. source
Read the paper
Topics
convergence geometric
← All NeurIPS 2025 Spotlight papers · Browse the whole archive