Exact computation of the Maximum Entropy Potential of spiking neural networks models

Abstract

Understanding how stimuli and synaptic connectivity in uence the statistics of spike patterns in neural networks is a central question in computational neuroscience. Maximum Entropy approach has been successfully used to characterize the statistical response of simultaneously recorded spiking neurons responding to stimuli. But, in spite of good performance in terms of prediction, the fitting parameters do not explain the underlying mechanistic causes of the observed correlations. On the other hand, mathematical models of spiking neurons (neuro-mimetic models) provide a probabilistic mapping between stimulus, network architecture and spike patterns in terms of conditional proba- bilities. In this paper we build an exact analytical mapping between neuro-mimetic and Maximum Entropy models.

arXiv admin note: text overlap with arXiv:1309.5873

References in corpus (2)

• Prediction of spatio-temporal patterns of neural activity from pairwise correlations
• A discrete time neural network model with spiking neurons II. Dynamics with noise

Cited by in corpus (7)

• Linear response in neuronal networks: from neurons dynamics to collective response
• A generalization of the maximum entropy principle for curved statistical manifolds
• Large Deviations Properties of Maximum Entropy Markov Chains from Spike Trains
• Thermodynamic Formalism in Neuronal Dynamics and Spike Train Statistics
• Linear response for spiking neuronal networks with unbounded memory
• Consistent model selection for estimating functional interactions among stochastic neurons with variable-length memory
• Information Entropy Production of Spatio-Temporal Maximum Entropy Distributions