We investigate storage capacity and generalization ability for two types of fully connected layered neural networks with non-monotonic transfer functions; random patterns are embedded into the networks by a Hebbian learning rule. One of them is a layered network in which a non-monotonic transfer...

In this paper, a new approximate formula to probability integral is deduced using theoretical analysis combining with computer numerical simulation. The absolute storage capacity of the Hopfield neural network is analyzed with this approximate formula and a more strict result is obtained.

A self-control mechanism for the dynamics of a three-state fully connected neural network is studied through the introduction of a time-dependent threshold. The self-adapting threshold is a function of both the neural and the pattern activity in the network. The time evolution of the order...

We consider a neural network of Stuart–Landau oscillators as an associative memory. This oscillator network with N elements is a system of an N-dimensional differential equation, works as an attractor neural network, and is expected to have no Lyapunov functions. Therefore, the technique of...

We investigate storage capacity of two types of fully connected layered neural networks with sparse coding when binary patterns are embedded into the networks by a Hebbian learning rule. One of them is a layered network, in which a transfer function of even layers is different from that of odd...

In this paper, the storage capacity of the Q-state complex phasor neural network is analysed with the signal-to-noise theory. The results indicate that the storage capacity of the model approaches that of the Hopfield model if the number Q is small; while the storage capacity is proportional to...

We investigate storage capacity of a fully connected layered neural network with Q(⩾2)-states clock neurons, including Q=∞ (corresponding to oscillatory neurons) and with intra-layer connections, where random Q-values patterns are embedded into the network by the Hebbian learning rule. We...

Highly oscillatory phenomena are omnipresent in applications. Two important underlying sources are stochastic fluctuations and deterministic randomness. In this paper, we will present heuristics, theorems, and a few illustrations on the Fourier spectrum and deterministic random time series,...