A note on solving nonlinear optimization problems in variable precision
arXiv:1812.03467
Abstract
This short note considers an efficient variant of the trust-region algorithm with dynamic accuracy proposed Carter (1993) and Conn, Gould and Toint (2000) as a tool for very high-performance computing, an area where it is critical to allow multi-precision computations for keeping the energy dissipation under control. Numerical experiments are presented indicating that the use of the considered method can bring substantial savings in objective function's and gradient's evaluation "energy costs" by efficiently exploiting multi-precision computations.
11 pages, 2 figures
References in corpus (4)
- Training Deep Neural Networks with 8-bit Floating Point Numbers
- Newton-Type Methods for Non-Convex Optimization Under Inexact Hessian Information
- Doing Moore with Less -- Leapfrogging Moore's Law with Inexactness for Supercomputing
- Adaptive Regularization Algorithms with Inexact Evaluations for Nonconvex Optimization
Cited by in corpus (3)
- Inexact Restoration for Minimization with Inexact Evaluation both of the Objective Function and the Constraints
- Minimization of nonsmooth nonconvex functions using inexact evaluations and its worst-case complexity
- A Stochastic Objective-Function-Free Adaptive Regularization Method with Optimal Complexity