Predictive regulation of neurotransmitter receptor pool for weight correction restriction

Artificial neural networks lack the adaptive properties of natural neural systems. The main goal of this work was to propose a theoretical model that would integrate different levels of homeostasis. In this paper, we propose a bio-inspired model, including homeostatic synaptic plasticity, limited by neurotransmitter receptors. The use of the echo state network for the regulation of neuron hyperparameters is suggested. The resulting model can be used to construct neural networks that perform adaptive tasks.