Finite Connectivity Attractor Neural Networks

Abstract

We study a family of diluted attractor neural networks with a finite average number of (symmetric) connections per neuron. As in finite connectivity spin glasses, their equilibrium properties are described by order parameter functions, for which we derive an integral equation in replica symmetric (RS) approximation. A bifurcation analysis of this equation reveals the locations of the paramagnetic to recall and paramagnetic to spin-glass transition lines in the phase diagram. The line separating the retrieval phase from the spin-glass phase is calculated at zero temperature. All phase transitions are found to be continuous.

17 pages, 4 figures

References in corpus (6)

• Statistical mechanics of the random K-SAT model
• Optimisation problems and replica symmetry breaking in finite connectivity spin-glasses
• The Statistical Physics of Regular Low-Density Parity-Check Error-Correcting Codes
• Tricritical Points in Random Combinatorics: the (2+p)-SAT case
• One+Infinite Dimensional Attractor Neural Networks
• Attractor Modulation and Proliferation in 1+$\infty$ Dimensional Neural Networks

Cited by in corpus (38)

• Multitasking associative networks
• Entropies of complex networks with hierarchically constrained topologies
• Associative memory on a small-world neural network
• Restricted Boltzmann Machine, recent advances and mean-field theory
• Parallel dynamics of disordered Ising spin systems on finitely connected random graphs
• Replicated Transfer Matrix Analysis of Ising Spin Models on `Small World' Lattices
• Immune networks: multi-tasking capabilities near saturation
• Finitely connected vector spin systems with random matrix interactions
• Analytic solution of attractor neural networks on scale-free graphs
• The Little-Hopfield model on a Random Graph
• Cavity approach for real variables on diluted graphs and application to synchronization in small-world lattices
• Dynamical replica analysis of disordered Ising spin systems on finitely connected random graphs
• Non-Convex Multi-species Hopfield models
• Parallel processing in immune networks
• Inference and learning in sparse systems with multiple states
• Spin models on random graphs with controlled topologies beyond degree constraints
• Meta-stable states in the hierarchical Dyson model drive parallel processing in the hierarchical Hopfield network
• Network resilience against intelligent attacks constrained by degree dependent node removal cost
• Reconstructing the Hopfield network as an inverse Ising problem
• Phase transitions in the three-state Ising spin-glass model with finite connectivity
• Dynamic rewiring in small world networks
• Topology and Dynamics of Attractor Neural Networks: The Role of Loopiness
• Slowly evolving geometry in recurrent neural networks I: extreme dilution regime
• Transient Dynamics of Sparsely Connected Hopfield Neural Networks with Arbitrary Degree Distributions
• Analogue neural networks on correlated random graphs
• Graph's Topology and Free Energy of a Spin Model on the Graph
• Vitality of Neural Networks under Reoccurring Catastrophic Failures
• Message passing algorithms for the Hopfield network reconstruction: threshold behavior and limitation
• Slowly evolving random graphs II: Adaptive geometry in finite-connectivity Hopfield models
• Sparse Hopfield network reconstruction with ${\ell }_1$ regularization
• Branching random graph model of rough surfaces describes thermal properties of the effective molecular potential
• Gardner optimal capacity of the diluted Blume-Emery-Griffiths neural network
• Transient dynamics for sequence processing neural networks: effect of degree distributions
• Statistical Mechanics of Dilute Batch Minority Games with Random External Information
• Multiplicative versus additive noise in multi-state neural networks
• Blume-Emery-Griffiths model on Random Graphs
• Multitasking network with fast noise
• Mean-field theory of vector spin models on networks with arbitrary degree distributions