Fast rate coding in heterogeneous spiking networks
A network of spiking neurons is said to operate in the rate coding regime if its dynamics can be accurately described by a continuous rate network. But how can discrete spike-based communication between neurons give rise to continuous rate-based dynamics? Computational studies showing that heterogeneous spiking networks can approximate the dynamics of recurrent neural networks (RNNs), a special…
A spiking neural network can function using rate coding if its behavior can be effectively modeled by a continuous rate network. However, the conversion of discrete, spike-based communication into continuous rate-based dynamics is not immediately clear. Computational research has demonstrated that heterogeneous spiking networks can mimic the behavior of recurrent neural networks (RNNs), a specific type of rate network.
These studies typically rely on assumptions that spikes follow a Poisson distribution or that the network operates at a slow pace. The current study challenges these assumptions: heterogeneous networks composed of non-Poisson neurons can indeed produce fast rate-based dynamics. The researchers demonstrated that signal transmission in feedforward networks of integrate-and-fire-type neurons can be accurately described by a history-dependent rate network that remains asymptotically precise even as the network grows in size, despite the presence of heterogeneity.
This finding contradicts the limitations of linear-nonlinear rate networks. By constructing a model that explains fast rate coding in heterogeneous networks, this research provides a basis for expanding rate-based analyses of neural recordings beyond the current reliance on slow dynamics or Poisson rate codes.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
