English

Generalization learning in a perceptron with binary synapses

Disordered Systems and Neural Networks 2012-11-14 v1

Abstract

We consider the generalization problem for a perceptron with binary synapses, implementing the Stochastic Belief-Propagation-Inspired (SBPI) learning algorithm which we proposed earlier, and perform a mean-field calculation to obtain a differential equation which describes the behaviour of the device in the limit of a large number of synapses N. We show that the solving time of SBPI is of order N*sqrt(log(N)), while the similar, well-known clipped perceptron (CP) algorithm does not converge to a solution at all in the time frame we considered. The analysis gives some insight into the ongoing process and shows that, in this context, the SBPI algorithm is equivalent to a new, simpler algorithm, which only differs from the CP algorithm by the addition of a stochastic, unsupervised meta-plastic reinforcement process, whose rate of application must be less than sqrt(2/(\pi * N)) for the learning to be achieved effectively. The analytical results are confirmed by simulations.

Keywords

Cite

@article{arxiv.1211.3024,
  title  = {Generalization learning in a perceptron with binary synapses},
  author = {Carlo Baldassi},
  journal= {arXiv preprint arXiv:1211.3024},
  year   = {2012}
}

Comments

16 pages, 4 figures

R2 v1 2026-06-21T22:37:39.528Z