Geometry of learning : Visualizing the performance of neural network supervised training methods /

Míreanna Comhchosúla

• On the alleviation of the problem of local minima in back-propagation /
  le: Magoulast, G. D.
• On the acceleration of the backpropagation training method /
  le: Vrahatis, M. N.
• Systolic array exploitation of a neural network inherent parallelism solving the nearest neighbor problem /
  le: Bekakos, Michael P.
• Neural networks for NP-complete problems /
  le: Budinich, Marco.
• Object oriented and genetically selected neural networks in a heterogeneous distributed environment /
  le: Cresswell, I.