Coherent lightwave associative memory system that possesses a carrier-frequency-controlled behavior

Abstract

We propose a coherent lightwave neural network system whose behavior can be controlled by the lightwave carrier frequency. We utilize the dependence of homodyne interference on the lightwave frequency by choosing the optical path lengths of the signal and the reference to be slightly different. The synaptic weights consist both of the optical path difference and variable delay realized by a spatial light modulator. A frequency-dependent learning is performed by a novel complex-valued Hebbian rule associated with the carrier frequency. Experiments demonstrate that the associative memory recalls different complex-amplitude patterns according to the carrier frequency variation successfully.

Developmental Learning With Behavioral Mode Tuning by Carrier-Frequency Modulation in Coherent Neural Networks

“…We propose to use this as the mode-tuning parameter to expect an effective increase in the variation in behavior. That is to say, if we fix the parameter , the behavioral mode is also fixed, whereas if we release it free to move to an optimal point in a self-organizing manner, then the network may learn and process properly and efficiently with an optimal parameter set slightly different from the previous one [18]. Fig.…”

“…A set of the amplitude-phase-type neurons form a CNN with a carrier frequency that works as the mode parameter [18]. The neural connection weight is expressed by the connection amplitude (transparency)…”

“…Its behavior can be modulated in accordance with the change in carrier frequency. For example, we have constructed an associative memory whose memorizing and recalling behavior is tuned by carrier-frequency modulation [18].…”

“…The authors previously proposed optical coherent neural networks (CNNs) [16]- [18]. The theoretical basis of the CNN lies in the complex-valued neural networks [19]- [22] that deal with not only signal amplitude but also signal phase.…”

Developmental Learning Based on Coherent Neural Networks with Behavioral Mode Tuning by Carrier-Frequency Modulation

“…The authors previously proposed optical coherent neural networks (CNNs) [11], [12]. The theoretical basis of the CNN lies in the complex-valued neural networks [13], [14], [15] that deal with not only signal amplitude but also signal phase.…”