中文

On-line learning in a discrete state space

凝聚态物理 2007-05-23 v1

摘要

On-line learning of a rule given by an N-dimensional Ising perceptron, is considered for the case when the student is constrained to take values in a discrete state space of size LNL^N. For L=2 no on-line algorithm can achieve a finite overlap with the teacher in the thermodynamic limit. However, if LL is on the order of N\sqrt{N}, Hebbian learning does achieve a finite overlap.

引用

@article{arxiv.cond-mat/9705257,
  title  = {On-line learning in a discrete state space},
  author = {W. Kinzel and R. Urbanczik},
  journal= {arXiv preprint arXiv:cond-mat/9705257},
  year   = {2007}
}

备注

7 pages, 1 Figure, Latex, submitted to J.Phys.A