Brainwashing Random Asymmetric `Neural' Networks
نویسندگان
چکیده
An algorithm for synaptic modiication (plasticity) is described by which a recurrently connected network of neuron-like units can organize itself to produce a sequence of activation states that does not repeat itself for a very long time. During the self-organization stage, the connections between the units undergo non-Hebbian modiications, which tend to decorrelate the activity of the units, thereby lengthening the period of the cyclic modes inherent in the network. It is shown that the periodicity of the activity rises exponentially with the amount of exposure to this plasticity algorithm. Threshold is also a critical parameter in determining cycle lengths, as is the rate of decay of the elds that accumulate at silent units.
منابع مشابه
Gyroscope Random Drift Modeling, using Neural Networks, Fuzzy Neural and Traditional Time- series Methods
In this paper statistical and time series models are used for determining the random drift of a dynamically Tuned Gyroscope (DTG). This drift is compensated with optimal predictive transfer function. Also nonlinear neural-network and fuzzy-neural models are investigated for prediction and compensation of the random drift. Finally the different models are compared together and their advantages a...
متن کاملThreshold Noise as a Source of Volatility in Random Synchronous Asymmetric Neural Networks
We study the diversity of complex spatio-temporal patterns of random synchronous asymmetric neural networks (RSANNs). Specifically, we investigate the impact of noisy thresholds on network performance and find that there is a narrow and interesting region of noise parameters where RSANNs display specific features of behavior desired for rapidly ‘thinking’ systems: accessibility to a large set o...
متن کاملIncrease in Complexity in Random Neural Networks
We study trie dynamics of a discrete time, continuous state neural network with random asymmetric couplings and random thresholds. Trie evolution of trie neurons is given m the thermodynamic limit by a set of dynamic mean-field equations obtained by
متن کاملAttractors in fully asymmetric neural networks
The statistical properties of the length of the cycles and of the weights of the attraction basins in fully asymmetric neural networks (i.e. with completely uncorrelated synapses) are computed in the framework of the annealed approximation which we previously introduced for the study of Kauffman networks. Our results show that this model behaves essentially as a Random Map possessing a reversal...
متن کاملEigenvalue spectra of asymmetric random matrices for multicomponent neural networks.
This paper focuses on large neural networks whose synaptic connectivity matrices are randomly chosen from certain random matrix ensembles. The dynamics of these networks can be characterized by the eigenvalue spectra of their connectivity matrices. In reality, neurons in a network do not necessarily behave in a similar way, but may belong to several different categories. The first study of the ...
متن کامل