A comparison of neural network training algorithms

Kristina A. Richardson, Robert F. Stengel

Research output: Contribution to conferencePaperpeer-review


Six methods of training neural networks to represent aerodynamic data are compared. Prior research has shown that the extended Kalman filter provides speed and accuracy advantages over the back-propagation method, and further quickening of training is desirable. We investigate four filter-based approaches (standard filter without process "noise", setting lower bounds on estimate-error covariance, periodically re-initializing the estimate-error covariance, and adding fictitious "process noise") and compare them to approaches incorporating genetic algorithms. A genetic algorithm uses a global search of the network weight space to identify feasible weights. Here, the algorithm is used independently and as a starting step for an extended Kalman filter. While the initial global search is intuitively appealing and may have value in some applications, the randomly initialized additive-noise filter is found to be the fastest training method in the present study.

Original languageEnglish (US)
Number of pages7
StatePublished - 1997
Event22nd Atmospheric Flight Mechanics Conference, 1997 - New Orleans, United States
Duration: Aug 11 1997Aug 13 1997


Other22nd Atmospheric Flight Mechanics Conference, 1997
Country/TerritoryUnited States
CityNew Orleans

All Science Journal Classification (ASJC) codes

  • Energy(all)
  • Aerospace Engineering


Dive into the research topics of 'A comparison of neural network training algorithms'. Together they form a unique fingerprint.

Cite this