Restart procedures for the conjugate gradient method

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

SGDR: Stochastic Gradient Descent with Warm Restarts (2016, methodology): “This scheme resembles the one of Powell (1977) for the conjugate gradient method.”
A SURVEY OF NONLINEAR CONJUGATE GRADIENT METHODS (2005, methodology): “Powell [ 88 ] proved that for a general nonlinear function, if (a) the step size sk = xk+1 xk approaches zero, (b) the line search is exact, and (c) the Lipschitz Assumption holds, then the PRP method is globally convergent.”
On Restart Procedures for the Conjugate Gradient Method (2004, methodology): “Motivated by an idea in [16], a new restart technique for the conjugate gradient method is proposed in this paper.”
Hierarchical Multinomial Processing Tree Models: A Latent-Trait Approach (2006, methodology): “The preliminary parameter estimates that emerge are used as the inital values for a conjugate-gradient algorithm ( Powell, 1977 ) maximizing the likelihood of the data with a relatively liberal stopping criterion.”
Data Assimilation for Numerical Weather Prediction: A Review (2009, unknown): “• The Powell C-G algorithm with restarts[294].”