Slowly evolving geometry in recurrent neural networks I: extreme dilution regime

Details Slowly evolving geometry in recurrent neural networks I: extreme dilution regime Slowly evolving geometry in recurrent neural networks I: extreme dilution regime

2004-03-24

arxiv.org/abs/cond-mat/0403615v1

J. Phys. A: Math. Gen. 37 (2004) 7653

Physics

Condensed Matter

Disordered Systems and Neural Networks

18 pages, 6 figures

Scientific paper

10.1088/0305-4470/37/31/002

We study extremely diluted spin models of neural networks in which the connectivity evolves in time, although adiabatically slowly compared to the neurons, according to stochastic equations which on average aim to reduce frustration. The (fast) neurons and (slow) connectivity variables equilibrate separately, but at different temperatures. Our model is exactly solvable in equilibrium. We obtain phase diagrams upon making the condensed ansatz (i.e. recall of one pattern). These show that, as the connectivity temperature is lowered, the volume of the retrieval phase diverges and the fraction of mis-aligned spins is reduced. Still one always retains a region in the retrieval phase where recall states other than the one corresponding to the `condensed' pattern are locally stable, so the associative memory character of our model is preserved.

Affiliated with

Also associated with

No associations

Rating

Slowly evolving geometry in recurrent neural networks I: extreme dilution regime 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 Slowly evolving geometry in recurrent neural networks I: extreme dilution regime, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Slowly evolving geometry in recurrent neural networks I: extreme dilution regime will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-89527