Storing sequences in binary tournament-based neural networks

TL;DR

This paper introduces an extension to clique-based neural networks that efficiently stores sequences using oriented connections and redundancy, incorporating anticipation and a double-layered structure for accurate retrieval, while maintaining biological plausibility.

Contribution

It presents a novel extension enabling efficient sequence storage in neural networks with biological plausibility and a combined hetero- and auto-association structure.

Findings

Enhanced sequence storage efficiency demonstrated

Increased biological plausibility of the model

Effective sequence retrieval with the proposed structure

Abstract

An extension to a recently introduced architecture of clique-based neural networks is presented. This extension makes it possible to store sequences with high efficiency. To obtain this property, network connections are provided with orientation and with flexible redundancy carried by both spatial and temporal redundancy, a mechanism of anticipation being introduced in the model. In addition to the sequence storage with high efficiency, this new scheme also offers biological plausibility. In order to achieve accurate sequence retrieval, a double layered structure combining hetero-association and auto-association is also proposed.