The Quasi-Quadratic Gradient is proposed as a new search direction that multiplies the BFGS inverse-Hessian approximation by the gradient to accelerate convergence over standard BFGS.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
citing papers explorer
-
Quasi-Quadratic Gradient: A New Direction for Accelerating the BFGS Method in Quasi-Newton Optimization
The Quasi-Quadratic Gradient is proposed as a new search direction that multiplies the BFGS inverse-Hessian approximation by the gradient to accelerate convergence over standard BFGS.
- Privacy-Preserving Logistic Regression Training with A Faster Gradient Variant