A theoretical framework for learning through structural plasticity

TL;DR

This paper develops a theoretical framework for understanding learning via structural plasticity in neural networks, incorporating biological features and analyzing effects like pruning and reorganization.

Contribution

It introduces a mean-field approach to model learning through structural plasticity, accounting for various biological network features and comparing theoretical predictions with simulations.

Findings

Framework captures effects of pruning and reorganization.

Quantifies learning and memory capabilities under different conditions.

Matches simulation results with theoretical predictions.

Abstract

A growing body of research indicates that structural plasticity mechanisms are crucial for learning and memory consolidation. Starting from a simple phenomenological model, we exploit a mean-field approach to develop a theoretical framework of learning through this kind of plasticity, capable of taking into account several features of the connectivity and pattern of activity of biological neural networks, including probability distributions of neuron firing rates, selectivity of the responses of single neurons to multiple stimuli, probabilistic connection rules and noisy stimuli. More importantly, it describes the effects of stabilization, pruning and reorganization of synaptic connections. This framework is used to compute the values of some relevant quantities used to characterize the learning and memory capabilities of the neuronal network in training and testing procedures as the number of training patterns and other model parameters vary. The results are then compared with those obtained through simulations with firing-rate-based neuronal network models.