INFERNO: A Novel Architecture for Generating Long Neuronal Sequences with Spikes

Abstract

Abstract. Human working memory is capable to generate dynamically robust and flexible neuronal sequences for action planning, problem solving and decision making. However, current neurocomputational models of working memory find hard to achieve these capabilities since intrinsic noise is difficult to stabilize over time and destroys global synchrony. As part of the principle of free-energy minimization proposed by Karl Friston, we propose a novel neural architecture to optimize the free-energy inherent to spiking recurrent neural networks to regulate their activity. We show for the first time that it is possible to stabilize iteratively the long-range control of a recurrent spiking neurons network over long sequences. We identify our architecture as the working memory composed by the Basal Ganglia and the Intra-Parietal Lobe for action selection and we make some comparisons with other networks such as deep neural networks and neural Turing machines. We name our architecture INFERNO for Iterative Free-Energy Optimization for Recurrent Neural Network. abstract environment.

Decoding Spiking Mechanism with Dynamic Learning on Neuron Population

“…Recently, deep neural networks have been used in cognitive neuroscience and primarily validated its feasibility in parametrizing nonlinear mapping from latent space to spike rates. One basic idea is fitting a recurrent neural network for sequential analysis [2], [18]. A similar way is to use convolutional neural networks to find relationship between stimulus and neuron response [14].…”