A memory optimal BFGS neural network training algorithm

McLoone, Sean and Asirvadam , Vijanth Sagayan (2002) A memory optimal BFGS neural network training algorithm. In: International Joint Conference in Neural Networks 2002, 12-17 May 2002, Honolulu, HI , USA .

Full text not available from this repository.
Official URL: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumb...


This paper considers the implementation of a novel memory optimal neural network training algorithm which maximises performance in relation to available memory. Mathematically, it is similar to the full memory BFGS training when there are no constraints on memory and to the variable memory (VM) BFGS when memory is limited. However, it requires less computations per iteration than VM and uses a much better strategy for discarding old curvature information when memory is limited

Item Type: Conference or Workshop Item (Paper)
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments / MOR / COE: Departments > Electrical & Electronic Engineering
Depositing User: Dr Vijanth Sagayan Asirvadam
Date Deposited: 12 Jan 2011 03:35
Last Modified: 12 Jan 2011 03:35
URI: http://scholars.utp.edu.my/id/eprint/3914

Actions (login required)

View Item
View Item