Chaotic Quaternionic Associative Memory

In this paper, we propose a chaotic quaternionic neuron model and a Chaotic Quaternionic Associative Memory (CQAM). The proposed chaotic quaternionic neuron model is based on the chaotic neuron model and the quaternionic neuron model. In the chaotic quaternionic neuron model, if the parameters are set appropriately, chaotic response can be generated. The proposed Chaotic Quaternionic Associative Memory is composed of chaotic quaternionic neuron models, and has a structure which is similar to the Hopfield network. In the proposed Chaotic Quaternionic Associative Memory, plural patterns are given to the network as external inputs at the same time, each pattern can be recalled separately. The proposed Chaotic Quaternionic Associative Memory makes use of the dynamic association ability of the chaotic quaternionic neuron model in order to realize pattern separation. We carried out a series of computer experiments and confirmed that (1) the chaotic quaternionic neuron can generate chaotic response when the parameters are set appropriately and (2) the pattern separation can be realized in the Chaotic Quaternionic Associative Memory.