BAUER, K and KREY, U (1991) ON THE STORAGE CAPACITY FOR TEMPORAL PATTERN SEQUENCES IN NETWORKS WITH DELAYS. ZEITSCHRIFT FUR PHYSIK B-CONDENSED MATTER, 84 (1). pp. 131-141. ISSN 0722-3277,
Full text not available from this repository.Abstract
We study the storage capacity for temporal sequences of random patterns in networks with arbitrary delays, evolving under parallel dynamics. For sequences with a common period T, made up of patterns which remain constant for DELTA-time steps, couplings with delays tau = DELTA.k - 1, where k is integer, are particularly important since they are tuned to the "rhythm" of the sequences. For networks with tuned delays only, we calculate the optimal storage capacity along the lines of Gardner [1] and find identical results to corresponding static cases, whereas untuned couplings induce several complications. For DELTA = 2, we consider networks with finite fractions 1 - a of untuned couplings, additionally weighted in strength by a parameter lambda with respect to the tuned couplings. For lambda-2(1 - a) << 1 we already find a pronounced decrease of the optimal storage capacity compared to the network where the fraction (1 - a) of untuned connections was cut. Thus for optimal error-free storage, the untuned couplings should be switched off. On the other hand, if errors are allowed and the couplings are chosen by a Hebbian prescription, the untuned couplings turn out to be useful, if the fraction a of tuned couplings exceeds a certain critical value, and the weight parameter lambda can then be optimized with respect to the storage capacity.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | NEURAL NETWORKS; CORRELATED PATTERNS; DYNAMIC OBJECTS; MODELS; RECOGNITION; |
| Depositing User: | Dr. Gernot Deinzer |
| Last Modified: | 19 Oct 2022 08:47 |
| URI: | https://pred.uni-regensburg.de/id/eprint/55174 |
Actions (login required)
![]() |
View Item |

