Quantized neural networks (QNNs) are widely used to achieve computationally efficient solutions recognition problems. Overall, eight-bit QNNs have almost the same accuracy as full-precision networks, but working several times faster. However, with lower quantization levels demonstrate inferior in comparison their classical analogs. To solve this issue, a number of quantization-aware training (Q...

Abstract An artificial neural network (ANN) has neurons that are interconnected. Inside the neuron, there is an activation function determines value issued by neuron. There several types of functions. The choice will determine result ANN. This paper describes performance comparison functions used in ANN processing sinusoidal signals. three being compared, those Sigmoid, Tansig, and ReLU. datase...

There has recently been a great deal of interest in inferring network connectivity from the spike trains in populations of neurons. One class of useful models that can be fit easily to spiking data is based on generalized linear point process models from statistics. Once the parameters for these models are fit, the analyst is left with a nonlinear spiking network model with delays, which in gen...

In this paper, we consider the problem of constructing a correction algorithm with aim increasing adaptive properties ΣΠ-neuron, relying solely on structure ΣΠ-neuron itself. To build corrector, logical-algebraic method data analysis is used. Comparison advantages neural network approach and suggests that combined to organization improves its efficiency allows one set rules reveal hidden patter...

Cover Illustration: The 3D reconstruction of super-resolution immunofluorescent confocal image shows that glutamate transporter 1 (GLT1; green)-labeled astrocytic processes enwrap neurons labeled by microtubule-associated protein 2 (MAP2, red; nuclei, blue) in primary mouse cortical glia-neuron mix culture. close proximity MAP2-GLT1 indicates the neuron-astrocyte interaction for maintaining hom...