Optimal pruning in neural networks.

@article{Barbato2000OptimalPI,
  title={Optimal pruning in neural networks.},
  author={D. M. L. Barbato and Osame Kinouchi},
  journal={Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics},
  year={2000},
  volume={62 6 Pt B},
  pages={
          8387-94
        }
}
We study pruning strategies in simple perceptrons subjected to supervised learning. Our analytical results, obtained through the statistical mechanics approach to learning theory, are independent of the learning algorithm used in the training process. We calculate the post-training distribution P(J) of synaptic weights, which depends only on the overlap rho(0) achieved by the learning algorithm before pruning and the fraction kappa of relevant weights in the teacher network. From this… CONTINUE READING