The capacity of \(q\)-state Potts neural networks with parallel retrieval dynamics (Q2467386): Difference between revisions
From MaRDI portal
Removed claim: author (P16): Item:Q638306 |
Changed an Item |
||
Property / author | |||
Property / author: Matthias Loewe / rank | |||
Normal rank |
Revision as of 02:21, 20 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The capacity of \(q\)-state Potts neural networks with parallel retrieval dynamics |
scientific article |
Statements
The capacity of \(q\)-state Potts neural networks with parallel retrieval dynamics (English)
0 references
21 January 2008
0 references
The so-called \(q\)-Potts neural networks are an extensions of Hopfield NN (neural network) to the case of \(q>=2\) states. The \(M\) fundamental memories are considered as N-dimensional i.i.d. random variables with probability \(1/q\). Storage and retrieval issues are discussed. Bounds for the storage capacity are shown.
0 references
Potts neural network
0 references
storage capacity
0 references
Hopfield model
0 references