Sign Up to like & get
recommendations!
1
Published in 2017 at "Neural Processing Letters"
DOI: 10.1007/s11063-017-9581-y
Abstract: Four gradient-based recurrent neural networks for computing the Drazin inverse of a square real matrix are developed. Theoretical analysis shows that any monotonically-increasing odd activation function ensures the global convergence performance of defined neural network…
read more here.
Keywords:
neural network;
drazin inverse;
computing drazin;
recurrent neural ... See more keywords