DOI: 10.1101/475319Nov 20, 2018Paper

Weight statistics controls dynamics in recurrent neural networks

BioRxiv : the Preprint Server for Biology
Patrick KraussClaus Metzner


Recurrent neural networks are complex non-linear systems, capable of ongoing activity in the absence of driving inputs. The dynamical properties of these systems, in particular their long-time attractor states, are determined on the microscopic level by the connection strengths $w\_{ij}$ between the individual neurons. However, little is known to which extent network dynamics is tunable on a more coarse-grained level by the {\em statistical} features of the weight matrix. In this work, we investigate the dynamical impact of three statistical parameters: {\em density} (the fraction of non-zero connections), {\em balance} (the ratio of excitatory to inhibitory connections), and {\em symmetry} (the fraction of neuron pairs with $w\_{ij}=w_{ji}$). By computing a 'phase diagram' of network dynamics, we find that balance is the essential control parameter: Its gradual increase from negative to positive values drives the system from oscillatory behavior into a chaotic regime, and eventually into stationary fix points. Only directly at the border of the chaotic regime do the neural networks display rich but regular dynamics, thus enabling actual information processing. These results suggest that the brain, too, is fine-tuned to the 'ed...Continue Reading

Related Concepts

Extracellular Matrix
Neural Stem Cells
PHLPP1 gene

Related Feeds

BioRxiv & MedRxiv Preprints

BioRxiv and MedRxiv are the preprint servers for biology and health sciences respectively, operated by Cold Spring Harbor Laboratory. Here are the latest preprint articles (which are not peer-reviewed) from BioRxiv and MedRxiv.

Related Papers

Physical Review. E, Statistical, Nonlinear, and Soft Matter Physics
E Greenfield, H Lecar
Physical Review. B, Condensed Matter
J u n q i n g LiJ i n n a n Gu
© 2021 Meta ULC. All rights reserved