GP-FL estimates the global Hessian at the server with Gaussian process regression on recent noisy gradient differences, claiming linear-quadratic convergence in over-the-air federated learning, but the required approximation property is assumed, not derived.
Training neural networks on remote edge devices for unseen class classification,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
GP-FL: Model-Based Hessian Estimation for Second-Order Over-the-Air Federated Learning
GP-FL estimates the global Hessian at the server with Gaussian process regression on recent noisy gradient differences, claiming linear-quadratic convergence in over-the-air federated learning, but the required approximation property is assumed, not derived.