Threshold Noise as a Source of Volatility in Random Synchronous Asymmetric Neural Networks

1997/12/12 by Henrik Bohr, Patrick McGuire, Bohr, Henrik +5 · 1 voice
Biochemistry, Genetics and Molecular Biology · Physics and Astronomy · #Adaptation and Self-Organizing Systems (nlin.AO) #Biological Physics (physics.bio-ph) #Disordered Systems and Neural Networks (cond-mat.dis-nn) #FOS: Biological sciences #FOS: Physical sciences #Neurons and Cognition (q-bio.NC) #adap-org #cond-mat.dis-nn #nlin.AO #physics.bio-ph #q-bio.NC

paper · pdf · doi:10.48550/arxiv.cond-mat/9712132

17 pages, 11 figures, submitted to Neural Computation

Abstract

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 of distinct, complex patterns.

Discussions

• The origin of "Brain Dynamics at the Edge of Chaos" is seen in this work https://arxiv.org/abs/cond-mat/0202190 https://www.sciencedirect.com/science/article/abs/pii/S0893608002000874 "Threshold disor [stackexchange, 0 points]

Related

• Threshold Disorder as a Source of Diverse and Complex Behavior in Random Nets
• Maximal Information Transfer and Behavior Diversity in Random Threshold Networks
• Stochastic resonance in three-neuron motifs
• Optimal rate-variance coding due to firing threshold adaptation near criticality
• Adaptive thresholds for layered neural networks with synaptic noise