conftrace_

compressed gradient descent

2 papers

Explore in graph

Also known as

CGD

Co-occurring keywords · papers mentioning both

distributed optimization (2) matrix stepsize (1) layer-wise compression (1) accelerated gradient descent (1) variance reduction (1) non-convex optimization (1) federated learning (1)

Papers

Det-CGD: Compressed Gradient Descent with Matrix Stepsizes for Non-Convex Optimization ICLR 2024 Acceleration for Compressed Gradient Descent in Distributed and Federated Optimization ICML 2020