The paper deals with transformation algorithms for decomposing Boolean neural networks (NNs); this being done with a view to possible VLSI implementation of NNs using threshold gates (TGs). We detail a possible tree decomposition for COMPARISON, and show how this can be used for the decomposition of Boolean functions (BFs) belonging to FN,m – the class of BFs of N variables that have exactly m groups of ones. Complexity estimates are given, the results being: (i) linear size; (ii) logarithmic depth; (iii) constant weights; and (iv) constant threshold.
|Title of host publication||European Symposium on Artificial Neural Networks|
|Publication status||Published - Apr 7 1993|
|Event||ESANN'93 - Brussels, Belgium|
Duration: Apr 7 1993 → …
|Period||4/7/93 → …|