Bayesian reconstruction of memories stored in neural networks from their connectivity

TL;DR

This paper investigates the theoretical and practical feasibility of reconstructing stored memories in neural networks from their connectivity matrices, introducing a Bayesian inference algorithm inspired by statistical physics.

Contribution

It provides a novel Bayesian inference algorithm for memory reconstruction in neural networks and analyzes its performance across different models, advancing connectomics research.

Findings

The algorithm performs comparably to PCA in certain models

Reconstruction success depends on network parameters and stored pattern complexity

Theoretical analysis clarifies when memory reconstruction is feasible

Abstract

The advent of comprehensive synaptic wiring diagrams of large neural circuits has created the field of connectomics and given rise to a number of open research questions. One such question is whether it is possible to reconstruct the information stored in a recurrent network of neurons, given its synaptic connectivity matrix. Here, we address this question by determining when solving such an inference problem is theoretically possible in specific attractor network models and by providing a practical algorithm to do so. The algorithm builds on ideas from statistical physics to perform approximate Bayesian inference and is amenable to exact analysis. We study its performance on three different models, compare the algorithm to standard algorithms such as PCA, and explore the limitations of reconstructing stored patterns from synaptic connectivity.