Abstract
In simulations on the Little-Hopfield model it is found that the convergence time to a stable state close to one of the embedded patterns scales like c(m0)logN10, where N is the size of the network and c(m0) depends on the initial macroscopic overlap m0 with the pattern. In a related model known as the pseudoinverse model the convergence time to the pattern is much smaller than log10N. The results are compared with other possible pattern recognition methods.
| Original language | English |
|---|---|
| Pages (from-to) | 2611-2614 |
| Number of pages | 4 |
| Journal | Physical Review A |
| Volume | 40 |
| Issue number | 5 |
| DOIs | |
| State | Published - 1989 |
| Externally published | Yes |