Conclusion

This article describes the operation of an oscillating neural network in which the information is encoded in the phase of the oscillating neurons, with the aim of reducing its energy consumption.

Some possible applications of this type of network have been demonstrated, such as image recognition or solving difficult optimization problems. Other applications in different fields are currently being evaluated, such as audio signal processing. A future development will be to perform online learning on ONNs, i.e. to include the ability to modify the neural network weights in real time according to new training data.

Finally, with regard to the environmental aspect, the low power consumption of oscillating neural networks may be a promising solution for the future of embedded artificial intelligence.