Summary
International Symposium on Nonlinear Theory and its Applications
2009
Session Number:C2L-A
Session:
Number:C2L-A2
Effects of Asymmetric and Symmetric STDP on the Topology of a Recurrent Neural Network Model
Hayato Fujita, Osamu Araki,
pp.-
Publication Date:2009/10/18
Online ISSN:2188-5079
Summary:
A simulation experiment showed that temporally asymmetric STDP leads to self-organization of feed-forward networks starting from a pacemaker neuron, by contrast, symmetric STDP does not[Masuda et al. 2007]. However, how the topology of a network without the pacemaker neuron is changed by the STDP is not well understood. The purpose of this study is to clarify how the topology of the network without a pacemaker neuron is changed by asymmetric or symmetric STDP. We executed numerical experiments and analyzed the topology of synaptic connections using some measures. The results suggest that asymmetric STDP turns a recurrent network into a feedforward network while symmetric STDP keeps the recurrent structure.