On the storage capacity of Hopfield models with correlated patterns
It is well known that the Hopfield model of a fully connected neural network with \(N\) neurons with the original Hebbian learning rule can store a number of patters equal to \(C N/\ln N\) without error, and equal to \(\alpha_c N\) (with \(\alpha_c\approx 0.138\) according to numerical and replica results) if a small fraction of errors is tolerated. These results hold with probability tending to one rapidly if the patterns are assumed to be i.i.d. random variables with each component a symmetric i.i.d. Rademacher random variable. The present paper presents a rigorous investigation or the question what becomes of these results if the patterns are correlated, either ``semantically or ``spatially. More specifically, it is supposed that either for any \(i\), \(\xi^\mu_i\), \(\mu\in {\mathbb N}\), form a Markov chain with state space \(\{-1,+1\}\), which are in turn independent for different \(i\), or vice versa that for each \(\mu\), \(\xi^\mu_i,i\in {\mathbb N}\), is an family of independent Markov chains. The results obtained are the following: 1) If perfect storage is required, the storage capacity decreases linearly the degree of correlation increases in the semantic case, while the opposite is true in the case of spatial correlation. 2) If storage with error is tolerated, the bounds obtained for \(a_c\) decrease with the correlation in both cases.
- A central limit theorem for the overlap in the Hopfield model
- An almost sure large deviation principle for the Hopfield model
- Gibbs measures and phase transitions
- Gibbs states of the Hopfield model in the regime of perfect memory
- Gibbs states of the Hopfield model with extensively many patterns.
- scientific article; zbMATH DE number 125396 (Why is no real title available?)
- scientific article; zbMATH DE number 563663 (Why is no real title available?)
- scientific article; zbMATH DE number 1096556 (Why is no real title available?)
- scientific article; zbMATH DE number 837115 (Why is no real title available?)
- Large deviation principles for the Hopfield model and the Kac-Hopfield model
- Lower bounds on the restitution error in the Hopfield model
- Neural network design for efficient information retrieval
- Neural networks and physical systems with emergent collective computational abilities
- Properties of neural networks storing spatially correlated patterns
- Rigorous bounds on the storage capacity of the dilute Hopfield model
- Rigorous results for the Hopfield model with many patterns
- Storage of sets of correlated data in neural network memories
- The capacity of the Hopfield associative memory
- The retrieval phase of the Hopfield model: A rigorous analysis of the overlap distribution
- Growth of correlation in the Hopfield model
- Moderate deviations for the overlap parameter in the Hopfield model
- Storage capacity of the exponential correlation associative memory
- Storing, learning and retrieving biased patterns
- On a model of associative memory with huge storage capacity
- Fluctuations of the free energy in the high temperature Hopfield model
- The storage capacity of the Hopfield model and moderate deviations
- Capacity of an associative memory model on random graph architectures
- Effect of spatial correlations on Hopfield neural network and dense associative memories
- Storage of correlated patterns in a perceptron
- A comparative study of sparse associative memories
- The path-integral analysis of an associative memory model storing an infinite number of finite limit cycles
- Storage of hierarchically correlated patterns
- Increased storage capacity for hierarchically structured information in a neural network of Ising type
- The relativistic Hopfield model with correlated patterns
- Modified pseudo-inverse neural networks storing correlated patterns
- Properties of neural networks storing spatially correlated patterns
- scientific article; zbMATH DE number 1275406 (Why is no real title available?)
- Storage of sets of correlated data in neural network memories
- scientific article; zbMATH DE number 592672 (Why is no real title available?)
- scientific article; zbMATH DE number 1121927 (Why is no real title available?)
- scientific article; zbMATH DE number 1159248 (Why is no real title available?)
- Topological pattern selection in recurrent networks
- Storing cycles in Hopfield-type networks with pseudoinverse learning rule: admissibility and network topology
- Retrieval Properties of Hopfield and Correlated Attractors in an Associative Memory Model
- Storage capacity of a neural network with state-dependent synapses
- On the storage capacity of generalized Hopfield models
- scientific article; zbMATH DE number 5022035 (Why is no real title available?)
- The Hopfield model on a sparse Erdös-Renyi graph
- The space of interactions in neural network models
- Capacity bounds for the CDMA system and a neural network: a moderate deviations approach
- scientific article; zbMATH DE number 5660968 (Why is no real title available?)
- Hopfield model with planted patterns: a teacher-student self-supervised learning model
- A spectral approach to Hebbian-like neural networks
- Binary associative memory networks: a review of mathematical framework and capacity analysis
- Opinion formation in a heterogeneous population --- a new approach to the Hopfield model
- The swapping algorithm for the Hopfield model with two patterns
- On the capacity of an associative memory model based on neural cliques
This page was built for publication: On the storage capacity of Hopfield models with correlated patterns
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1296722)
