conftrace_

2022 NeurIPS NeurIPS 2022

Local Linear Convergence of Gradient Methods for Subspace Optimization via Strict Complementarity

Abstract

We consider optimization problems in which the goal is to find a $k$-dimensional subspace of $\mathbb{R}^n$, $k<

Authors

Ron Fisher, Dan Garber

Topics

Mathematics & Optimization > Optimization > Continuous Optimization Mathematics & Optimization > Optimization > Theory

Keywords

linear convergence gradient method local convergence subspace optimization strict complementarity

More from NeurIPS 2022

All NeurIPS 2022 papers →

Generalised Implicit Neural Representations

EZNAS: Evolving Zero-Cost Proxies For Neural Architecture Scoring

Personalized Federated Learning towards Communication Efficiency, Robustness and Fairness

Joint Learning of 2D-3D Weakly Supervised Semantic Segmentation

RNNs of RNNs: Recursive Construction of Stable Assemblies of Recurrent Neural Networks

Training Spiking Neural Networks with Event-driven Backpropagation