Related Experiment Videos

Improved backpropagation learning in neural networks with windowed momentum

Ernest Istook1, Tony Martinez

  • 1Brigham Young University, Computer Science Department, 3361 TMCB, P.O. Box 26576, Provo Utah 84602-6576, USA. butch@axon.cs.byu.edu

Summary

Windowed Momentum, a new neural network training algorithm, significantly speeds up convergence time by using recent weight updates. This method achieves an average 32% speedup while maintaining or improving accuracy, even with varied data presentation orders.

Related Concept Videos