Diagonal Approximation of the Hessian by Finite Differences for Unconstrained Optimization

2020 ◽  
Vol 185 (3) ◽  
pp. 859-879
Author(s):  
Neculai Andrei
2013 ◽  
Vol 2013 ◽  
pp. 1-5 ◽  
Author(s):  
Mahboubeh Farid ◽  
Wah June Leong ◽  
Najmeh Malekmohammadi ◽  
Mustafa Mamat

We present a new gradient method that uses scaling and extra updating within the diagonal updating for solving unconstrained optimization problem. The new method is in the frame of Barzilai and Borwein (BB) method, except that the Hessian matrix is approximated by a diagonal matrix rather than the multiple of identity matrix in the BB method. The main idea is to design a new diagonal updating scheme that incorporates scaling to instantly reduce the large eigenvalues of diagonal approximation and otherwise employs extra updates to increase small eigenvalues. These approaches give us a rapid control in the eigenvalues of the updating matrix and thus improve stepwise convergence. We show that our method is globally convergent. The effectiveness of the method is evaluated by means of numerical comparison with the BB method and its variant.


Author(s):  
Lisiane Trevisan ◽  
Juliane Donadel ◽  
Bianca de Castro
Keyword(s):  

Kerntechnik ◽  
2010 ◽  
Vol 75 (4) ◽  
pp. 210-213 ◽  
Author(s):  
D. Suescún Díaz ◽  
A. Senra Martinez
Keyword(s):  

10.37236/24 ◽  
2002 ◽  
Vol 1000 ◽  
Author(s):  
A. Di Bucchianico ◽  
D. Loeb

We survey the mathematical literature on umbral calculus (otherwise known as the calculus of finite differences) from its roots in the 19th century (and earlier) as a set of “magic rules” for lowering and raising indices, through its rebirth in the 1970’s as Rota’s school set it on a firm logical foundation using operator methods, to the current state of the art with numerous generalizations and applications. The survey itself is complemented by a fairly complete bibliography (over 500 references) which we expect to update regularly.


Sign in / Sign up

Export Citation Format

Share Document