Analytic solution of attractor neural networks on scale-free graphs

DOI10.1088/0305-4470/37/37/002MaRDI QIDQ4822784zbMATH OpenOpenAlexFDO

Authors Isaac Pérez Castillo, B. Wemmenhove, J. P. L. Hatchett, A. C. C. Coolen, Nikos S. Skantzos, T. Nikoletopoulos

Publication date 25 October 2004

Published in Journal of Physics A: Mathematical and General (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/cond-mat/0404018

Mathematics Subject Classification ID

Dynamical systems in biology (37N25) Neural networks for/in biological studies, artificial life and related topics (92B20) Neural nets applied to problems in time-dependent statistical mechanics (82C32)

Abstract: We study the influence of network topology on retrieval properties of recurrent neural networks, using replica techniques for diluted systems. The theory is presented for a network with an arbitrary degree distribution $p(k)$ and applied to power law distributions $p(k)sim{k}^{-gamma}$, i.e. to neural networks on scale-free graphs. A bifurcation analysis identifies phase boundaries between the paramagnetic phase and either a retrieval phase or a spin glass phase. Using a population dynamics algorithm, the retrieval overlap and spin glass order parameters may be calculated throughout the phase diagram. It is shown that there is an enhancement of the retrieval properties compared with a Poissonian random graph. We compare our findings with simulations.

Recommendations

• Finite connectivity attractor neural networks
• Retrieval properties of diluted attractor neural networks
• Retrieval phase diagrams for attractor neural networks with optimal interactions
• Analogue neural networks on correlated random graphs
• Spontaneous scale-free structure of spike flow graphs in recurrent neural networks

Cited in

(9)

• Feed-forward chains of recurrent attractor neural networks near saturation
• Attractor modulation and proliferation in \((1+\infty)\)-dimensional neural networks
• Equilibrium statistical mechanics on correlated random graphs
• Thermodynamics of bidirectional associative memories
• Sparse connection density underlies the maximal functional difference between random and scale-free networks
• Hierarchical neural networks perform both serial and parallel processing
• Finite connectivity attractor neural networks
• Information and Topology in Attractor Neural Networks
• Spontaneous scale-free structure of spike flow graphs in recurrent neural networks

This page was built for publication: Analytic solution of attractor neural networks on scale-free graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4822784)