Enhancing neural-network performance via assortativity

Details Enhancing neural-network performance via assortativity Enhancing neural-network performance via assortativity

2010-12-08

arxiv.org/abs/1012.1813v1

Physics

Condensed Matter

Disordered Systems and Neural Networks

9 pages, 7 figures

Scientific paper

The performance of attractor neural networks has been shown to depend crucially on the heterogeneity of the underlying topology. We take this analysis a step further by examining the effect of degree-degree correlations -- or assortativity -- on neural-network behavior. We make use of a method recently put forward for studying correlated networks and dynamics thereon, both analytically and computationally, which is independent of how the topology may have evolved. We show how the robustness to noise is greatly enhanced in assortative (positively correlated) neural networks, especially if it is the hub neurons that store the information.

Affiliated with

Also associated with

No associations

Rating

Enhancing neural-network performance via assortativity does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Enhancing neural-network performance via assortativity, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Enhancing neural-network performance via assortativity will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-558858