Heterogeneity in spiking coding networks

This project is a collaboration between the Boris Gutkin, Sophie Denève and myself at the Group for Neural Theory (ENS, Paris).

It is widely accepted that neurons do not form a homogeneous population, but that there is large variability between neurons. Therefore, the way in which individual neurons respond to a stimulus (their encoding properties or receptive field) also varies. A classical example is the difference between ’type 1′ and ’type 2′ neurons. Apart from these intrinsic ‘encoding’ properties of neurons, the ‘decoding’ properties of neurons also show a large variability. Surprisingly however, neural coding models often treat neural networks as homogeneous populations of neurons. Here we show that networks with spiking neurons with heterogeneous encoding and decoding properties can do optimal online stimulus representation.

In a collaboration with Sander Keemink, Raimon Bullich is linking the results we found in the single neuron experiments to network effects in this framework. Stefan Iacob is now researching the effects of decay heterogeneity in these networks, and how the slow synapses that are needed to compensate for a drift in error representation can be learned. 


 

link to manuscript:
Zeldenrust, F., Gutkin, B., & Denéve, S. (2021). Efficient and robust coding in heterogeneous recurrent networks. PLOS Computational Biology, 17(4), e1008673. https://doi.org/10.1371/journal.pcbi.1008673 (preprint)
link to conference abstracts:

Cosyne 2013: Zeldenrust, Denève, Gutkin – 2013 – Matching encoding and decoding with spiking neurons

Stefan Iacob’s poster at SNUFA 2026

link to code:  GitHub, Open Source Brain