Title: Scaling of neural algorithms

  • Objective: how to charcterize complexity of different neural network methods?
  • Participants: W. Duch
  • Time: since 1996
  • Results: scaling of the number of adaptive parameters with the number of classes/features investigated and a few simple structures found; one technical report and a conference paper written. 
  • Open problems: how to charcterize data to determine which methods will work? Mak some computer experiments to see if it is possible to find the simplest structures found for MLP, LR and other netowrks by optimization of network architectures.

    • Links to other Duch-Lab projects, many talks and to other papers on various subjects.

    Working log (local accsess only), maintained by Wlodzislaw Duch.